From 7f982bb0591ea5f34e72ce8893f319fc0df0fead Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 14:07:05 -0800
Subject: [PATCH 01/26] .github/workflows/ci-conda-known-test-failures.json:
 Add random failure https://github.com/sagemath/sage/issues/36832

---
 .github/workflows/ci-conda-known-test-failures.json | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 2d828ac98fb..4680c5e7c81 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -47,6 +47,9 @@
     "sage.rings.polynomial.skew_polynomial_finite_field": {
         "failed": true
     },
+    "sage.schemes.elliptic_curves.ell_field": {
+        "failed": true
+    },
     "sage.tests.gap_packages": {
         "failed": true
     }

From 48cde676c8b0a94df9d40f6fa56ebc43227694a7 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 13:38:10 -0800
Subject: [PATCH 02/26] src/sage/doctest/forker.py: Show '# [failed in
 baseline]' earlier

---
 src/sage/doctest/forker.py | 41 +++++++++++++++++++++++++++++++-------
 1 file changed, 34 insertions(+), 7 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index 4315cd1ba9f..12b4bdfaa80 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -513,7 +513,7 @@ def __init__(self, *args, **kwds):
 
         - ``stdout`` -- an open file to restore for debugging
 
-        - ``checker`` -- None, or an instance of
+        - ``checker`` -- ``None``, or an instance of
           :class:`doctest.OutputChecker`
 
         - ``verbose`` -- boolean, determines whether verbose printing
@@ -522,6 +522,9 @@ def __init__(self, *args, **kwds):
         - ``optionflags`` -- Controls the comparison with the expected
           output.  See :mod:`testmod` for more information.
 
+        - ``baseline`` -- ``None`` or a dictionary, the ``baseline_stats``
+          value
+
         EXAMPLES::
 
             sage: from sage.doctest.parsing import SageOutputChecker
@@ -535,6 +538,9 @@ def __init__(self, *args, **kwds):
         O = kwds.pop('outtmpfile', None)
         self.msgfile = kwds.pop('msgfile', None)
         self.options = kwds.pop('sage_options')
+        self.baseline = kwds.pop('baseline', None)
+        if self.baseline is None:
+            self.baseline = {}
         doctest.DocTestRunner.__init__(self, *args, **kwds)
         self._fakeout = SageSpoofInOut(O)
         if self.msgfile is None:
@@ -1721,12 +1727,20 @@ def serial_dispatch(self):
         """
         for source in self.controller.sources:
             heading = self.controller.reporter.report_head(source)
+            basename = source.basename
+            if self.controller.baseline_stats:
+                the_baseline_stats = self.controller.baseline_stats.get(basename, {})
+            else:
+                the_baseline_stats = {}
+            if the_baseline_stats.get('failed', False):
+                heading += "  # [failed in baseline]"
             if not self.controller.options.only_errors:
                 self.controller.log(heading)
 
             with tempfile.TemporaryFile() as outtmpfile:
                 result = DocTestTask(source)(self.controller.options,
-                                             outtmpfile, self.controller.logger)
+                                             outtmpfile, self.controller.logger,
+                                             baseline=the_baseline_stats)
                 outtmpfile.seek(0)
                 output = bytes_to_str(outtmpfile.read())
 
@@ -1984,10 +1998,17 @@ def sel_exit():
                             # Start a new worker.
                             import copy
                             worker_options = copy.copy(opt)
+                            basename = source.basename
+                            if self.controller.baseline_stats:
+                                the_baseline_stats = self.controller.baseline_stats.get(basename, {})
+                            else:
+                                the_baseline_stats = {}
                             if target_endtime is not None:
                                 worker_options.target_walltime = (target_endtime - now) / (max(1, pending_tests / opt.nthreads))
-                            w = DocTestWorker(source, options=worker_options, funclist=[sel_exit])
+                            w = DocTestWorker(source, options=worker_options, baseline=the_baseline_stats, funclist=[sel_exit])
                             heading = self.controller.reporter.report_head(w.source)
+                            if the_baseline_stats.get('failed', False):
+                                heading += "  # [failed in baseline]"
                             if not self.controller.options.only_errors:
                                 w.messages = heading + "\n"
                             # Store length of heading to detect if the
@@ -2146,7 +2167,7 @@ class should be accessed by the child process.
         sage: reporter.report(FDS, False, W.exitcode, result, "")
             [... tests, ... s]
     """
-    def __init__(self, source, options, funclist=[]):
+    def __init__(self, source, options, funclist=[], baseline=None):
         """
         Initialization.
 
@@ -2170,6 +2191,7 @@ def __init__(self, source, options, funclist=[]):
         self.source = source
         self.options = options
         self.funclist = funclist
+        self.baseline = baseline
 
         # Open pipe for messages. These are raw file descriptors,
         # not Python file objects!
@@ -2241,7 +2263,7 @@ def run(self):
         os.close(self.rmessages)
         msgpipe = os.fdopen(self.wmessages, "w")
         try:
-            task(self.options, self.outtmpfile, msgpipe, self.result_queue)
+            task(self.options, self.outtmpfile, msgpipe, self.result_queue, baseline=self.baseline)
         finally:
             msgpipe.close()
             self.outtmpfile.close()
@@ -2507,7 +2529,8 @@ def __init__(self, source):
         """
         self.source = source
 
-    def __call__(self, options, outtmpfile=None, msgfile=None, result_queue=None):
+    def __call__(self, options, outtmpfile=None, msgfile=None, result_queue=None, *,
+                 baseline=None):
         """
         Calling the task does the actual work of running the doctests.
 
@@ -2524,6 +2547,9 @@ def __call__(self, options, outtmpfile=None, msgfile=None, result_queue=None):
         - ``result_queue`` -- an instance of :class:`multiprocessing.Queue`
           to store the doctest result. For testing, this can also be None.
 
+        - ``baseline`` -- ``None`` or a dictionary, the ``baseline_stats``
+          value.
+
         OUTPUT:
 
         - ``(doctests, result_dict)`` where ``doctests`` is the number of
@@ -2559,7 +2585,8 @@ def __call__(self, options, outtmpfile=None, msgfile=None, result_queue=None):
                 outtmpfile=outtmpfile,
                 msgfile=msgfile,
                 sage_options=options,
-                optionflags=doctest.NORMALIZE_WHITESPACE | doctest.ELLIPSIS)
+                optionflags=doctest.NORMALIZE_WHITESPACE | doctest.ELLIPSIS,
+                baseline=baseline)
             runner.basename = self.source.basename
             runner.filename = self.source.path
             N = options.file_iterations

From d52858cd55f394f8b9287ed74cc0d5935394c663 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 15:30:09 -0800
Subject: [PATCH 03/26] Refactor through new function
 DocTestController.source_baseline

---
 src/sage/doctest/control.py   | 28 ++++++++++++++++++++++++++++
 src/sage/doctest/forker.py    | 35 +++++++++++++----------------------
 src/sage/doctest/reporting.py | 17 +++++++----------
 3 files changed, 48 insertions(+), 32 deletions(-)

diff --git a/src/sage/doctest/control.py b/src/sage/doctest/control.py
index fd92dc386bf..37562e1717c 100644
--- a/src/sage/doctest/control.py
+++ b/src/sage/doctest/control.py
@@ -1097,6 +1097,34 @@ def sort_key(source):
                 return -self.stats.get(basename, default).get('walltime', 0), basename
             self.sources = sorted(self.sources, key=sort_key)
 
+    def source_baseline(self, source):
+        r"""
+        Return the ``baseline_stats`` value of ``source``.
+
+        INPUT:
+
+        - ``source`` -- a :class:`DocTestSource` instance
+
+        OUTPUT:
+
+        A dictionary.
+
+        EXAMPLES::
+
+            sage: from sage.doctest.control import DocTestDefaults, DocTestController
+            sage: from sage.env import SAGE_SRC
+            sage: import os
+            sage: filename = os.path.join(SAGE_SRC,'sage','doctest','util.py')
+            sage: DD = DocTestDefaults()
+            sage: DC = DocTestController(DD, [filename])
+            sage: DC.source_baseline(DC.sources[0])
+            {}
+        """
+        if self.baseline_stats:
+            basename = source.basename
+            return self.baseline_stats.get(basename, {})
+        return {}
+
     def run_doctests(self):
         """
         Actually runs the doctests.
diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index 12b4bdfaa80..cb8348e2732 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -522,8 +522,7 @@ def __init__(self, *args, **kwds):
         - ``optionflags`` -- Controls the comparison with the expected
           output.  See :mod:`testmod` for more information.
 
-        - ``baseline`` -- ``None`` or a dictionary, the ``baseline_stats``
-          value
+        - ``baseline`` -- dictionary, the ``baseline_stats`` value
 
         EXAMPLES::
 
@@ -538,9 +537,7 @@ def __init__(self, *args, **kwds):
         O = kwds.pop('outtmpfile', None)
         self.msgfile = kwds.pop('msgfile', None)
         self.options = kwds.pop('sage_options')
-        self.baseline = kwds.pop('baseline', None)
-        if self.baseline is None:
-            self.baseline = {}
+        self.baseline = kwds.pop('baseline', {})
         doctest.DocTestRunner.__init__(self, *args, **kwds)
         self._fakeout = SageSpoofInOut(O)
         if self.msgfile is None:
@@ -1727,12 +1724,8 @@ def serial_dispatch(self):
         """
         for source in self.controller.sources:
             heading = self.controller.reporter.report_head(source)
-            basename = source.basename
-            if self.controller.baseline_stats:
-                the_baseline_stats = self.controller.baseline_stats.get(basename, {})
-            else:
-                the_baseline_stats = {}
-            if the_baseline_stats.get('failed', False):
+            baseline = self.controller.source_baseline(source)
+            if baseline.get('failed', False):
                 heading += "  # [failed in baseline]"
             if not self.controller.options.only_errors:
                 self.controller.log(heading)
@@ -1740,7 +1733,7 @@ def serial_dispatch(self):
             with tempfile.TemporaryFile() as outtmpfile:
                 result = DocTestTask(source)(self.controller.options,
                                              outtmpfile, self.controller.logger,
-                                             baseline=the_baseline_stats)
+                                             baseline=baseline)
                 outtmpfile.seek(0)
                 output = bytes_to_str(outtmpfile.read())
 
@@ -1998,16 +1991,12 @@ def sel_exit():
                             # Start a new worker.
                             import copy
                             worker_options = copy.copy(opt)
-                            basename = source.basename
-                            if self.controller.baseline_stats:
-                                the_baseline_stats = self.controller.baseline_stats.get(basename, {})
-                            else:
-                                the_baseline_stats = {}
+                            baseline = self.controller.source_baseline(source)
                             if target_endtime is not None:
                                 worker_options.target_walltime = (target_endtime - now) / (max(1, pending_tests / opt.nthreads))
-                            w = DocTestWorker(source, options=worker_options, baseline=the_baseline_stats, funclist=[sel_exit])
+                            w = DocTestWorker(source, options=worker_options, funclist=[sel_exit], baseline=baseline)
                             heading = self.controller.reporter.report_head(w.source)
-                            if the_baseline_stats.get('failed', False):
+                            if baseline.get('failed', False):
                                 heading += "  # [failed in baseline]"
                             if not self.controller.options.only_errors:
                                 w.messages = heading + "\n"
@@ -2148,6 +2137,8 @@ class should be accessed by the child process.
     - ``funclist`` -- a list of callables to be called at the start of
       the child process.
 
+    - ``baseline`` -- dictionary, the ``baseline_stats`` value
+
     EXAMPLES::
 
         sage: from sage.doctest.forker import DocTestWorker, DocTestTask
@@ -2263,7 +2254,8 @@ def run(self):
         os.close(self.rmessages)
         msgpipe = os.fdopen(self.wmessages, "w")
         try:
-            task(self.options, self.outtmpfile, msgpipe, self.result_queue, baseline=self.baseline)
+            task(self.options, self.outtmpfile, msgpipe, self.result_queue,
+                 baseline=self.baseline)
         finally:
             msgpipe.close()
             self.outtmpfile.close()
@@ -2547,8 +2539,7 @@ def __call__(self, options, outtmpfile=None, msgfile=None, result_queue=None, *,
         - ``result_queue`` -- an instance of :class:`multiprocessing.Queue`
           to store the doctest result. For testing, this can also be None.
 
-        - ``baseline`` -- ``None`` or a dictionary, the ``baseline_stats``
-          value.
+        - ``baseline`` -- a dictionary, the ``baseline_stats`` value.
 
         OUTPUT:
 
diff --git a/src/sage/doctest/reporting.py b/src/sage/doctest/reporting.py
index a86153ce326..dcc0dee792c 100644
--- a/src/sage/doctest/reporting.py
+++ b/src/sage/doctest/reporting.py
@@ -399,10 +399,7 @@ def report(self, source, timeout, return_code, results, output, pid=None):
             postscript = self.postscript
             stats = self.stats
             basename = source.basename
-            if self.controller.baseline_stats:
-                the_baseline_stats = self.controller.baseline_stats.get(basename, {})
-            else:
-                the_baseline_stats = {}
+            baseline = self.controller.source_baseline(source)
             cmd = self.report_head(source)
             try:
                 ntests, result_dict = results
@@ -423,14 +420,14 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                         fail_msg += " (and interrupt failed)"
                     else:
                         fail_msg += " (with %s after interrupt)" % signal_name(sig)
-                if the_baseline_stats.get('failed', False):
+                if baseline.get('failed', False):
                     fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s timed out:" % (fail_msg, "*"*70, process_name))
                 log(output)
                 log("*"*70)
                 postscript['lines'].append(cmd + "  # %s" % fail_msg)
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
-                if not the_baseline_stats.get('failed', False):
+                if not baseline.get('failed', False):
                     self.error_status |= 4
             elif return_code:
                 if return_code > 0:
@@ -439,14 +436,14 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     fail_msg = "Killed due to %s" % signal_name(-return_code)
                 if ntests > 0:
                     fail_msg += " after testing finished"
-                if the_baseline_stats.get('failed', False):
+                if baseline.get('failed', False):
                     fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s failed:" % (fail_msg,"*"*70, process_name))
                 log(output)
                 log("*"*70)
                 postscript['lines'].append(cmd + "  # %s" % fail_msg)
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
-                if not the_baseline_stats.get('failed', False):
+                if not baseline.get('failed', False):
                     self.error_status |= (8 if return_code > 0 else 16)
             else:
                 if hasattr(result_dict, 'walltime') and hasattr(result_dict.walltime, '__len__') and len(result_dict.walltime) > 0:
@@ -509,10 +506,10 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     f = result_dict.failures
                     if f:
                         fail_msg = "%s failed" % (count_noun(f, "doctest"))
-                        if the_baseline_stats.get('failed', False):
+                        if baseline.get('failed', False):
                             fail_msg += " [failed in baseline]"
                         postscript['lines'].append(cmd + "  # %s" % fail_msg)
-                        if not the_baseline_stats.get('failed', False):
+                        if not baseline.get('failed', False):
                             self.error_status |= 1
                     if f or result_dict.err == 'tab':
                         stats[basename] = {"failed": True, "walltime": wall, "ntests": ntests}

From 1849493abcd1231647a878ecd8be373ebd038f39 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 12:31:24 -0800
Subject: [PATCH 04/26] WIP

---
 .github/workflows/build.yml     |  80 +-----------------------
 .github/workflows/ci-conda.yml  | 104 --------------------------------
 src/bin/sage-runtests           |   2 +
 src/sage/combinat/words/word.py |   4 +-
 src/sage/doctest/control.py     |   1 +
 src/sage/doctest/forker.py      |  47 ++++++++++++---
 src/sage_setup/clean.py         |   2 +-
 7 files changed, 45 insertions(+), 195 deletions(-)
 delete mode 100644 .github/workflows/ci-conda.yml

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index c9282086422..82c2f19dc8f 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -96,39 +96,6 @@ jobs:
           MAKE: make -j2 --output-sync=recurse
           SAGE_NUM_THREADS: 2
 
-      - name: Build modularized distributions
-        if: (success() || failure()) && steps.worktree.outcome == 'success'
-        run: make V=0 tox && make SAGE_CHECK=no pypi-wheels
-        working-directory: ./worktree-image
-        env:
-          MAKE: make -j2 --output-sync=recurse
-          SAGE_NUM_THREADS: 2
-
-      - name: Static code check with pyright
-        if: (success() || failure()) && steps.worktree.outcome == 'success'
-        uses: jakebailey/pyright-action@v1
-        with:
-          version: 1.1.332
-          # Many warnings issued by pyright are not yet helpful because there is not yet enough type information.
-          no-comments: true
-          working-directory: ./worktree-image
-        env:
-          # To avoid out of memory errors
-          NODE_OPTIONS: --max-old-space-size=8192
-      
-      - name: Static code check with pyright (annotated)
-        if: (success() || failure()) && steps.worktree.outcome == 'success'
-        uses: jakebailey/pyright-action@v1
-        with:
-          version: 1.1.332
-          # Issue errors
-          no-comments: false
-          level: error
-          working-directory: ./worktree-image
-        env:
-          # To avoid out of memory errors
-          NODE_OPTIONS: --max-old-space-size=8192
-
       - name: Clean (fallback to non-incremental)
         id: clean
         if: (success() || failure()) && steps.worktree.outcome == 'success' && steps.incremental.outcome != 'success'
@@ -154,52 +121,9 @@ jobs:
 
       # Testing
 
-      - name: Test changed files (sage -t --new)
-        if: (success() || failure()) && steps.build.outcome == 'success'
-        run: |
-          # We run tests with "sage -t --new"; this only tests the uncommitted changes.
-          ./sage -t --new -p2
-        working-directory: ./worktree-image
-        env:
-          MAKE: make -j2 --output-sync=recurse
-          SAGE_NUM_THREADS: 2
-
-      - name: Test modularized distributions
-        if: (success() || failure()) && steps.build.outcome == 'success'
-        run: make V=0 tox && make pypi-wheels-check
-        working-directory: ./worktree-image
-        env:
-          MAKE: make -j2 --output-sync=recurse
-          SAGE_NUM_THREADS: 2
-
-      - name: Pytest
-        if: contains(github.ref, 'pytest')
-        run: |
-          ../sage -python -m pip install coverage pytest-xdist
-          ../sage -python -m coverage run -m pytest -c tox.ini --doctest-modules || true
-        working-directory: ./worktree-image/src
-        env:
-          # Increase the length of the lines in the "short summary"
-          COLUMNS: 120
-
       - name: Test all files (sage -t --all --long)
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
-          ../sage -python -m pip install coverage
-          ../sage -python -m coverage run ./bin/sage-runtests --all --long -p2 --random-seed=286735480429121101562228604801325644303
-        working-directory: ./worktree-image/src
-
-      - name: Prepare coverage results
-        if: (success() || failure()) && steps.build.outcome == 'success'
-        run: |
-          ./venv/bin/python3 -m coverage combine src/.coverage/
-          ./venv/bin/python3 -m coverage xml
-          mkdir -p coverage-report
-          mv coverage.xml coverage-report/
+          ./sage -python -m pip install coverage
+          ./sage -python -m coverage run ./src/bin/sage-runtests --format github --baseline-stats-path=.github/workflows/ci-conda-known-test-failures.json -p4 --random-seed=286735480429121101562228604801325644303 src/sage/combinat/words src/sage/plot/plot.py src/sage_setup/clean.py
         working-directory: ./worktree-image
-
-      - name: Upload coverage to codecov
-        if: (success() || failure()) && steps.build.outcome == 'success'
-        uses: codecov/codecov-action@v3
-        with:
-          directory: ./worktree-image/coverage-report
diff --git a/.github/workflows/ci-conda.yml b/.github/workflows/ci-conda.yml
deleted file mode 100644
index 6432b7a0a6a..00000000000
--- a/.github/workflows/ci-conda.yml
+++ /dev/null
@@ -1,104 +0,0 @@
-name: Build & Test using Conda
-
-on:
-  push:
-    tags:
-      - '*'
-    branches:
-      - 'public/build/**-runci'
-  pull_request:
-  workflow_dispatch:
-    # Allow to run manually
-
-concurrency:
-  # Cancel previous runs of this workflow for the same branch
-  group: ${{ github.workflow }}-${{ github.ref }}
-  cancel-in-progress: true
-
-jobs:
-  test:
-    name: Conda
-    runs-on: ${{ matrix.os }}
-
-    strategy:
-      matrix:
-        os: [ubuntu-latest, macos-latest]
-        python: ['3.9', '3.10', '3.11']
-        # Optional environment is disabled for now as its not yet working
-        # environment: [environment, environment-optional]
-        conda-env: [environment]
-
-    steps:
-      - uses: actions/checkout@v4
-
-      - name: Merge CI fixes from sagemath/sage
-        run: |
-          .ci/merge-fixes.sh
-        env:
-          GH_TOKEN: ${{ github.token }}
-          SAGE_CI_FIXES_FROM_REPOSITORIES: ${{ vars.SAGE_CI_FIXES_FROM_REPOSITORIES }}
-
-      - name: Create conda environment files
-        run: ./bootstrap-conda
-
-      - name: Cache conda packages
-        uses: actions/cache@v3
-        with:
-          path: ~/conda_pkgs_dir
-          key:
-            ${{ runner.os }}-conda-${{ hashFiles('src/environment-3.11.yml') }}
-
-      - name: Setup Conda environment
-        uses: conda-incubator/setup-miniconda@v2
-        with:
-          python-version: ${{ matrix.python }}
-          miniforge-version: latest
-          use-mamba: true
-          channels: conda-forge
-          channel-priority: true
-          activate-environment: sage
-          environment-file: src/${{ matrix.conda-env }}-${{ matrix.python }}.yml
-
-      - name: Print Conda environment
-        shell: bash -l {0}
-        run: |
-          conda info
-          conda list
-
-      - name: Configure
-        shell: bash -l {0}
-        continue-on-error: true
-        run: |
-          ./bootstrap
-          echo "::add-matcher::.github/workflows/configure-systempackage-problem-matcher.json"
-          ./configure --enable-build-as-root --with-python=$CONDA_PREFIX/bin/python --prefix=$CONDA_PREFIX --enable-system-site-packages $(for pkg in $(./sage -package list :standard: --has-file spkg-configure.m4 --has-file distros/conda.txt --exclude rpy2); do echo --with-system-$pkg=force; done)
-          echo "::remove-matcher owner=configure-system-package-warning::"
-          echo "::remove-matcher owner=configure-system-package-error::"
-
-      - name: Build
-        shell: bash -l {0}
-        run: |
-          # Use --no-deps and pip check below to verify that all necessary dependencies are installed via conda.
-          pip install --no-build-isolation --no-deps -v -v -e ./pkgs/sage-conf ./pkgs/sage-setup
-          pip install --no-build-isolation --no-deps --config-settings editable_mode=compat -v -v -e ./src
-        env:
-          SAGE_NUM_THREADS: 2
-        
-      - name: Verify dependencies
-        if: success() || failure()
-        shell: bash -l {0}
-        run: pip check
-
-      - name: Test
-        if: success() || failure()
-        shell: bash -l {0}
-        run: ./sage -t --all --baseline-stats-path=.github/workflows/ci-conda-known-test-failures.json -p0
-
-      - name: Print logs
-        if: always()
-        run: |
-          for file in $(find . -type f -name "*.log"); do
-              echo "::group::$file"
-              cat "$file"
-              echo "::endgroup::"
-          done
diff --git a/src/bin/sage-runtests b/src/bin/sage-runtests
index e25cebb48a8..25444933bf6 100755
--- a/src/bin/sage-runtests
+++ b/src/bin/sage-runtests
@@ -34,6 +34,8 @@ if __name__ == "__main__":
     what.add_argument("--installed", action="store_true", default=False, help="test all installed modules of the Sage library")
     parser.add_argument("--logfile", type=argparse.FileType('a'), metavar="FILE", help="log all output to FILE")
 
+    parser.add_argument("--format", choices=["sage", "github"], default="sage",
+                        help="set format of error messages and warnings")
     parser.add_argument("-l", "--long", action="store_true", default=False, help="include lines with the phrase 'long time'")
     parser.add_argument("-s", "--short", dest="target_walltime", nargs='?',
                         type=int, default=-1, const=300, metavar="SECONDS",
diff --git a/src/sage/combinat/words/word.py b/src/sage/combinat/words/word.py
index 132195589e9..240e66b6a24 100644
--- a/src/sage/combinat/words/word.py
+++ b/src/sage/combinat/words/word.py
@@ -148,7 +148,7 @@ def Word(data=None, alphabet=None, length=None, datatype=None, caching=True, RSK
 
     Word over a string with a parent::
 
-        sage: w = Word("abbabaab", alphabet="abc"); w
+        sage: w = Word("abbabaab", daslkdlsakdlksaldkasldklsakdlsak, alphabet="abc"); w
         word: abbabaab
         sage: w.parent()
         Finite words over {'a', 'b', 'c'}
@@ -253,7 +253,7 @@ class FiniteWord_char(WordDatatype_char, FiniteWord_class):
         True
 
         sage: w.is_square_free()
-        False
+        Falsish
         sage: w[:-1].is_square_free()
         True
 
diff --git a/src/sage/doctest/control.py b/src/sage/doctest/control.py
index 37562e1717c..01a9e080384 100644
--- a/src/sage/doctest/control.py
+++ b/src/sage/doctest/control.py
@@ -139,6 +139,7 @@ def __init__(self, **kwds):
         self.show_skipped = False
         self.target_walltime = -1
         self.baseline_stats_path = None
+        self.format = "sage"
 
         # sage-runtests contains more optional tags. Technically, adding
         # auto_optional_tags here is redundant, since that is added
diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index cb8348e2732..a6b43b37aa2 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1235,16 +1235,40 @@ def _failure_header(self, test, example, message='Failed example:'):
         """
         out = [self.DIVIDER]
         with OriginalSource(example):
-            if test.filename:
-                if test.lineno is not None and example.lineno is not None:
-                    lineno = test.lineno + example.lineno + 1
-                else:
-                    lineno = '?'
-                out.append('File "%s", line %s, in %s' %
-                           (test.filename, lineno, test.name))
-            else:
-                out.append('Line %s, in %s' % (example.lineno + 1, test.name))
-            out.append(message)
+            match self.options.format:
+                case 'sage':
+                    if test.filename:
+                        if test.lineno is not None and example.lineno is not None:
+                            lineno = test.lineno + example.lineno + 1
+                        else:
+                            lineno = '?'
+                        out.append('File "%s", line %s, in %s' %
+                                   (test.filename, lineno, test.name))
+                    else:
+                        out.append('Line %s, in %s' % (example.lineno + 1, test.name))
+                    out.append(message)
+                case 'github':
+                    # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
+                    if message.startswith('Warning: '):
+                        command = f'::warning title={message}'
+                        message = message[len('Warning: '):]
+                    elif self.baseline.get('failed', False):
+                        command = f'::notice title={message}'
+                        message += ' [failed in baseline]'
+                    else:
+                        command = f'::error title={message}'
+                    if extra := getattr(example, 'extra', None):
+                        message += f': {extra}'
+                    if test.filename:
+                        command += f',file={test.filename}'
+                        if test.lineno is not None and example.lineno is not None:
+                            lineno = test.lineno + example.lineno + 1
+                            command += f',line={lineno}'
+                        lineno = None
+                    else:
+                        command += f',line={example.lineno + 1}'
+                    command += f'::{message}'
+                    out.append(command)
             source = example.source
             out.append(doctest._indent(source))
             return '\n'.join(out)
@@ -1417,6 +1441,7 @@ def report_failure(self, out, test, example, got, globs):
         """
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
+            example.extra = f'Got: {got!r}'
             returnval = doctest.DocTestRunner.report_failure(self, out, test, example, got)
             if self.options.debug:
                 self._fakeout.stop_spoofing()
@@ -1567,6 +1592,8 @@ def report_unexpected_exception(self, out, test, example, exc_info):
         """
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
+
+            example.extra = "Exception raised: " + repr("".join(traceback.format_exception(*exc_info)))
             returnval = doctest.DocTestRunner.report_unexpected_exception(self, out, test, example, exc_info)
             if self.options.debug:
                 self._fakeout.stop_spoofing()
diff --git a/src/sage_setup/clean.py b/src/sage_setup/clean.py
index e9c81c9ed1a..bac2ed26417 100644
--- a/src/sage_setup/clean.py
+++ b/src/sage_setup/clean.py
@@ -101,7 +101,7 @@ def _find_stale_files(site_packages, python_packages, python_modules, ext_module
         sage: stale_iter = _find_stale_files(SAGE_LIB, python_packages, python_modules, [], extra_files)
         sage: from importlib.machinery import EXTENSION_SUFFIXES
         sage: skip_extensions = tuple(EXTENSION_SUFFIXES)
-        sage: for f in stale_iter:
+        sage: for f in stale_iter:dxfcgvhbjkl;,mkjnbhgvcfgvhbjnkml,
         ....:     if f.endswith(skip_extensions): continue
         ....:     if '/ext_data/' in f: continue
         ....:     print('Found stale file: ' + f)

From 8aca1adf8f09988c47533963851c1606756fbff8 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 17:46:37 -0800
Subject: [PATCH 05/26] sage -t: Add option '--format github'

---
 src/bin/sage-runtests       |  2 ++
 src/sage/doctest/control.py |  1 +
 src/sage/doctest/forker.py  | 47 +++++++++++++++++++++++++++++--------
 3 files changed, 40 insertions(+), 10 deletions(-)

diff --git a/src/bin/sage-runtests b/src/bin/sage-runtests
index e25cebb48a8..25444933bf6 100755
--- a/src/bin/sage-runtests
+++ b/src/bin/sage-runtests
@@ -34,6 +34,8 @@ if __name__ == "__main__":
     what.add_argument("--installed", action="store_true", default=False, help="test all installed modules of the Sage library")
     parser.add_argument("--logfile", type=argparse.FileType('a'), metavar="FILE", help="log all output to FILE")
 
+    parser.add_argument("--format", choices=["sage", "github"], default="sage",
+                        help="set format of error messages and warnings")
     parser.add_argument("-l", "--long", action="store_true", default=False, help="include lines with the phrase 'long time'")
     parser.add_argument("-s", "--short", dest="target_walltime", nargs='?',
                         type=int, default=-1, const=300, metavar="SECONDS",
diff --git a/src/sage/doctest/control.py b/src/sage/doctest/control.py
index 37562e1717c..01a9e080384 100644
--- a/src/sage/doctest/control.py
+++ b/src/sage/doctest/control.py
@@ -139,6 +139,7 @@ def __init__(self, **kwds):
         self.show_skipped = False
         self.target_walltime = -1
         self.baseline_stats_path = None
+        self.format = "sage"
 
         # sage-runtests contains more optional tags. Technically, adding
         # auto_optional_tags here is redundant, since that is added
diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index cb8348e2732..a6b43b37aa2 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1235,16 +1235,40 @@ def _failure_header(self, test, example, message='Failed example:'):
         """
         out = [self.DIVIDER]
         with OriginalSource(example):
-            if test.filename:
-                if test.lineno is not None and example.lineno is not None:
-                    lineno = test.lineno + example.lineno + 1
-                else:
-                    lineno = '?'
-                out.append('File "%s", line %s, in %s' %
-                           (test.filename, lineno, test.name))
-            else:
-                out.append('Line %s, in %s' % (example.lineno + 1, test.name))
-            out.append(message)
+            match self.options.format:
+                case 'sage':
+                    if test.filename:
+                        if test.lineno is not None and example.lineno is not None:
+                            lineno = test.lineno + example.lineno + 1
+                        else:
+                            lineno = '?'
+                        out.append('File "%s", line %s, in %s' %
+                                   (test.filename, lineno, test.name))
+                    else:
+                        out.append('Line %s, in %s' % (example.lineno + 1, test.name))
+                    out.append(message)
+                case 'github':
+                    # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
+                    if message.startswith('Warning: '):
+                        command = f'::warning title={message}'
+                        message = message[len('Warning: '):]
+                    elif self.baseline.get('failed', False):
+                        command = f'::notice title={message}'
+                        message += ' [failed in baseline]'
+                    else:
+                        command = f'::error title={message}'
+                    if extra := getattr(example, 'extra', None):
+                        message += f': {extra}'
+                    if test.filename:
+                        command += f',file={test.filename}'
+                        if test.lineno is not None and example.lineno is not None:
+                            lineno = test.lineno + example.lineno + 1
+                            command += f',line={lineno}'
+                        lineno = None
+                    else:
+                        command += f',line={example.lineno + 1}'
+                    command += f'::{message}'
+                    out.append(command)
             source = example.source
             out.append(doctest._indent(source))
             return '\n'.join(out)
@@ -1417,6 +1441,7 @@ def report_failure(self, out, test, example, got, globs):
         """
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
+            example.extra = f'Got: {got!r}'
             returnval = doctest.DocTestRunner.report_failure(self, out, test, example, got)
             if self.options.debug:
                 self._fakeout.stop_spoofing()
@@ -1567,6 +1592,8 @@ def report_unexpected_exception(self, out, test, example, exc_info):
         """
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
+
+            example.extra = "Exception raised: " + repr("".join(traceback.format_exception(*exc_info)))
             returnval = doctest.DocTestRunner.report_unexpected_exception(self, out, test, example, exc_info)
             if self.options.debug:
                 self._fakeout.stop_spoofing()

From 72ac0ece1e818ba637f28acbf3e8528d600d7f13 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 18:19:24 -0800
Subject: [PATCH 06/26] .github/workflows/build.yml: Use sage -t --format
 github

---
 .github/workflows/build.yml | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index c9282086422..2b87d6edb60 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -158,7 +158,7 @@ jobs:
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
           # We run tests with "sage -t --new"; this only tests the uncommitted changes.
-          ./sage -t --new -p2
+          ./sage -t --new -p2 --format github
         working-directory: ./worktree-image
         env:
           MAKE: make -j2 --output-sync=recurse
@@ -185,9 +185,9 @@ jobs:
       - name: Test all files (sage -t --all --long)
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
-          ../sage -python -m pip install coverage
-          ../sage -python -m coverage run ./bin/sage-runtests --all --long -p2 --random-seed=286735480429121101562228604801325644303
-        working-directory: ./worktree-image/src
+          ./sage -python -m pip install coverage
+          ./sage -python -m coverage run ./src/bin/sage-runtests --all --long -p2 --format github --random-seed=286735480429121101562228604801325644303
+        working-directory: ./worktree-image
 
       - name: Prepare coverage results
         if: (success() || failure()) && steps.build.outcome == 'success'

From 7a121e65c5bc2bb6b870dffb6c18d27dca52d6ea Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 20:45:27 -0800
Subject: [PATCH 07/26] src/sage/doctest/control.py: Fix doctest

---
 src/sage/doctest/control.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/sage/doctest/control.py b/src/sage/doctest/control.py
index 37562e1717c..96a73f600b2 100644
--- a/src/sage/doctest/control.py
+++ b/src/sage/doctest/control.py
@@ -1117,6 +1117,7 @@ def source_baseline(self, source):
             sage: filename = os.path.join(SAGE_SRC,'sage','doctest','util.py')
             sage: DD = DocTestDefaults()
             sage: DC = DocTestController(DD, [filename])
+            sage: DC.expand_files_into_sources()
             sage: DC.source_baseline(DC.sources[0])
             {}
         """

From 0bd0077e8a2bfc2e286d0aca270c362490e5b3c2 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 21:13:05 -0800
Subject: [PATCH 08/26] src/sage/doctest/control.py: Log 'Using
 --baseline-stats-path=...'

---
 src/sage/doctest/control.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/sage/doctest/control.py b/src/sage/doctest/control.py
index 96a73f600b2..1a128983187 100644
--- a/src/sage/doctest/control.py
+++ b/src/sage/doctest/control.py
@@ -1171,6 +1171,8 @@ def run_doctests(self):
             iterations = ", ".join(iterations)
             if iterations:
                 iterations = " (%s)" % (iterations)
+            if self.baseline_stats:
+                self.log(f"Using --baseline-stats-path={self.options.baseline_stats_path}")
             self.log("Doctesting %s%s%s." % (filestr, threads, iterations))
             self.reporter = DocTestReporter(self)
             self.dispatcher = DocTestDispatcher(self)

From 24779a92ba8a417575c700cc0c6b7ff2e601cf48 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 21:49:57 -0800
Subject: [PATCH 09/26] src/sage/doctest/forker.py: Rewrite without 'match'

---
 src/sage/doctest/forker.py | 67 +++++++++++++++++++-------------------
 1 file changed, 34 insertions(+), 33 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index a6b43b37aa2..88c006ca48f 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1235,40 +1235,41 @@ def _failure_header(self, test, example, message='Failed example:'):
         """
         out = [self.DIVIDER]
         with OriginalSource(example):
-            match self.options.format:
-                case 'sage':
-                    if test.filename:
-                        if test.lineno is not None and example.lineno is not None:
-                            lineno = test.lineno + example.lineno + 1
-                        else:
-                            lineno = '?'
-                        out.append('File "%s", line %s, in %s' %
-                                   (test.filename, lineno, test.name))
-                    else:
-                        out.append('Line %s, in %s' % (example.lineno + 1, test.name))
-                    out.append(message)
-                case 'github':
-                    # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
-                    if message.startswith('Warning: '):
-                        command = f'::warning title={message}'
-                        message = message[len('Warning: '):]
-                    elif self.baseline.get('failed', False):
-                        command = f'::notice title={message}'
-                        message += ' [failed in baseline]'
+            if self.options.format == 'sage':
+                if test.filename:
+                    if test.lineno is not None and example.lineno is not None:
+                        lineno = test.lineno + example.lineno + 1
                     else:
-                        command = f'::error title={message}'
-                    if extra := getattr(example, 'extra', None):
-                        message += f': {extra}'
-                    if test.filename:
-                        command += f',file={test.filename}'
-                        if test.lineno is not None and example.lineno is not None:
-                            lineno = test.lineno + example.lineno + 1
-                            command += f',line={lineno}'
-                        lineno = None
-                    else:
-                        command += f',line={example.lineno + 1}'
-                    command += f'::{message}'
-                    out.append(command)
+                        lineno = '?'
+                    out.append('File "%s", line %s, in %s' %
+                               (test.filename, lineno, test.name))
+                else:
+                    out.append('Line %s, in %s' % (example.lineno + 1, test.name))
+                out.append(message)
+            elif self.options.format == 'github':
+                # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
+                if message.startswith('Warning: '):
+                    command = f'::warning title={message}'
+                    message = message[len('Warning: '):]
+                elif self.baseline.get('failed', False):
+                    command = f'::notice title={message}'
+                    message += ' [failed in baseline]'
+                else:
+                    command = f'::error title={message}'
+                if extra := getattr(example, 'extra', None):
+                    message += f': {extra}'
+                if test.filename:
+                    command += f',file={test.filename}'
+                    if test.lineno is not None and example.lineno is not None:
+                        lineno = test.lineno + example.lineno + 1
+                        command += f',line={lineno}'
+                    lineno = None
+                else:
+                    command += f',line={example.lineno + 1}'
+                command += f'::{message}'
+                out.append(command)
+            else:
+                raise ValueError(f'unknown format option: {self.options.format}')
             source = example.source
             out.append(doctest._indent(source))
             return '\n'.join(out)

From ee0cb6b033f3d31be8c09cf8705a307c21e2b84c Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 23:03:28 -0800
Subject: [PATCH 10/26] .github/workflows/ci-conda-known-test-failures.json:
 Add https://github.com/sagemath/sage/issues/36939

---
 .github/workflows/ci-conda-known-test-failures.json | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 4680c5e7c81..18b898c6685 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -35,6 +35,9 @@
     "sage.modular.modform.l_series_gross_zagier": {
         "failed": true
     },
+    "sage.parallel.map_reduce": {
+        "failed": true
+    },
     "sage.rings.function_field.drinfeld_modules.morphism": {
         "failed": true
     },

From fed6570ef3dffebaa58689ddae78c514addcac16 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 23:15:09 -0800
Subject: [PATCH 11/26] .github/workflows/ci-conda-known-test-failures.json:
 Add https://github.com/sagemath/sage/issues/35973

---
 .github/workflows/ci-conda-known-test-failures.json | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 18b898c6685..8010bfc3727 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -53,6 +53,9 @@
     "sage.schemes.elliptic_curves.ell_field": {
         "failed": true
     },
+    "sage.structure.coerce_actions": {
+        "failed": true
+    },
     "sage.tests.gap_packages": {
         "failed": true
     }

From 22ef48640af4094cfad0f999b94645fdc5617b64 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 20 Dec 2023 23:20:05 -0800
Subject: [PATCH 12/26] .github/workflows/ci-conda-known-test-failures.json:
 Add some failures from https://github.com/sagemath/sage/pull/36372

---
 .github/workflows/ci-conda-known-test-failures.json | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 8010bfc3727..12ae507bedc 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -2,6 +2,9 @@
     "sage_setup.clean": {
         "failed": true
     },
+    "sage.algebras.fusion_rings.fusion_ring": {
+        "failed": true
+    },
     "sage.combinat.cluster_algebra_quiver.quiver": {
         "failed": true
     },
@@ -47,12 +50,18 @@
     "sage.rings.polynomial.multi_polynomial_libsingular": {
         "failed": true
     },
+    "sage.rings.polynomial.polynomial_element": {
+        "failed": true
+    },
     "sage.rings.polynomial.skew_polynomial_finite_field": {
         "failed": true
     },
     "sage.schemes.elliptic_curves.ell_field": {
         "failed": true
     },
+    "sage.sets.recursively_enumerated_set": {
+        "failed": true
+    },
     "sage.structure.coerce_actions": {
         "failed": true
     },

From 5fc07c7e3574c2a2e096862555e2da3fc5c9c816 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 10:48:40 -0800
Subject: [PATCH 13/26] src/bin/sage-runtests: Generalize argparse hack

---
 src/bin/sage-runtests | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/bin/sage-runtests b/src/bin/sage-runtests
index e25cebb48a8..4a8b05cfab6 100755
--- a/src/bin/sage-runtests
+++ b/src/bin/sage-runtests
@@ -141,7 +141,8 @@ if __name__ == "__main__":
             in_filenames = True
             new_arguments.append('--')
         new_arguments.append(arg)
-        afterlog = bool(arg == '--logfile')
+        afterlog = arg in ['--logfile', '--stats_path', '--stats-path',
+                           '--baseline_stats_path', '--baseline-stats-path']
 
     args = parser.parse_args(new_arguments)
 

From 93c4e7d83347c85205d455331eba469b9050e521 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 11:23:44 -0800
Subject: [PATCH 14/26] DocTestReporter.report_head: Move all printing of
 'failed in baseline' here

---
 src/sage/doctest/forker.py    |  4 ----
 src/sage/doctest/reporting.py | 37 ++++++++++++++++++++---------------
 2 files changed, 21 insertions(+), 20 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index cb8348e2732..a7379ec97bd 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1725,8 +1725,6 @@ def serial_dispatch(self):
         for source in self.controller.sources:
             heading = self.controller.reporter.report_head(source)
             baseline = self.controller.source_baseline(source)
-            if baseline.get('failed', False):
-                heading += "  # [failed in baseline]"
             if not self.controller.options.only_errors:
                 self.controller.log(heading)
 
@@ -1996,8 +1994,6 @@ def sel_exit():
                                 worker_options.target_walltime = (target_endtime - now) / (max(1, pending_tests / opt.nthreads))
                             w = DocTestWorker(source, options=worker_options, funclist=[sel_exit], baseline=baseline)
                             heading = self.controller.reporter.report_head(w.source)
-                            if baseline.get('failed', False):
-                                heading += "  # [failed in baseline]"
                             if not self.controller.options.only_errors:
                                 w.messages = heading + "\n"
                             # Store length of heading to detect if the
diff --git a/src/sage/doctest/reporting.py b/src/sage/doctest/reporting.py
index dcc0dee792c..7bb40ff4bd6 100644
--- a/src/sage/doctest/reporting.py
+++ b/src/sage/doctest/reporting.py
@@ -162,14 +162,16 @@ def were_doctests_with_optional_tag_run(self, tag):
             return True
         return False
 
-    def report_head(self, source):
+    def report_head(self, source, fail_msg=None):
         """
-        Return the "sage -t [options] file.py" line as string.
+        Return the ``sage -t [options] file.py`` line as string.
 
         INPUT:
 
         - ``source`` -- a source from :mod:`sage.doctest.sources`
 
+        - ``fail_msg`` -- ``None`` or a string
+
         EXAMPLES::
 
             sage: from sage.doctest.reporting import DocTestReporter
@@ -190,6 +192,8 @@ def report_head(self, source):
             sage: DD.long = True
             sage: print(DTR.report_head(FDS))
             sage -t --long .../sage/doctest/reporting.py
+            sage: print(DTR.report_head(FDS), "Failed by self-sabotage")
+            sage -t --long .../sage/doctest/reporting.py  # Failed by self-sabotage
         """
         cmd = "sage -t"
         if self.controller.options.long:
@@ -206,6 +210,13 @@ def report_head(self, source):
         if environment != "sage.repl.ipython_kernel.all_jupyter":
             cmd += f" --environment={environment}"
         cmd += " " + source.printpath
+        baseline = self.controller.source_baseline(source)
+        if fail_msg:
+            cmd += "  # " + fail_msg
+        if baseline.get('failed', False):
+            if not fail_msg:
+                cmd += "  #"
+            cmd += " [failed in baseline]"
         return cmd
 
     def report(self, source, timeout, return_code, results, output, pid=None):
@@ -420,12 +431,10 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                         fail_msg += " (and interrupt failed)"
                     else:
                         fail_msg += " (with %s after interrupt)" % signal_name(sig)
-                if baseline.get('failed', False):
-                    fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s timed out:" % (fail_msg, "*"*70, process_name))
                 log(output)
                 log("*"*70)
-                postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                postscript['lines'].append(self.report_head(source, fail_msg))
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
                 if not baseline.get('failed', False):
                     self.error_status |= 4
@@ -436,12 +445,10 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     fail_msg = "Killed due to %s" % signal_name(-return_code)
                 if ntests > 0:
                     fail_msg += " after testing finished"
-                if baseline.get('failed', False):
-                    fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s failed:" % (fail_msg,"*"*70, process_name))
                 log(output)
                 log("*"*70)
-                postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                postscript['lines'].append(self.report_head(source, fail_msg))
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
                 if not baseline.get('failed', False):
                     self.error_status |= (8 if return_code > 0 else 16)
@@ -458,13 +465,13 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     log("    Error in doctesting framework (bad result returned)\n%s\nTests run before error:" % ("*"*70))
                     log(output)
                     log("*"*70)
-                    postscript['lines'].append(cmd + "  # Testing error: bad result")
+                    postscript['lines'].append(self.report_head(source, "Testing error: bad result"))
                     self.error_status |= 64
                 elif result_dict.err == 'noresult':
                     log("    Error in doctesting framework (no result returned)\n%s\nTests run before error:" % ("*"*70))
                     log(output)
                     log("*"*70)
-                    postscript['lines'].append(cmd + "  # Testing error: no result")
+                    postscript['lines'].append(self.report_head(source, "Testing error: no result"))
                     self.error_status |= 64
                 elif result_dict.err == 'tab':
                     if len(result_dict.tab_linenos) > 5:
@@ -473,11 +480,11 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     if len(result_dict.tab_linenos) > 1:
                         tabs = "s" + tabs
                     log("    Error: TAB character found at line%s" % (tabs))
-                    postscript['lines'].append(cmd + "  # Tab character found")
+                    postscript['lines'].append(self.report_head(source, "Tab character found"))
                     self.error_status |= 32
                 elif result_dict.err == 'line_number':
                     log("    Error: Source line number found")
-                    postscript['lines'].append(cmd + "  # Source line number found")
+                    postscript['lines'].append(self.report_head(source, "Source line number found"))
                     self.error_status |= 256
                 elif result_dict.err is not None:
                     # This case should not occur
@@ -494,7 +501,7 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     if output:
                         log("Tests run before doctest exception:\n" + output)
                         log("*"*70)
-                    postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                    postscript['lines'].append(self.report_head(source, fail_msg))
                     if hasattr(result_dict, 'tb'):
                         log(result_dict.tb)
                     if hasattr(result_dict, 'walltime'):
@@ -506,9 +513,7 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     f = result_dict.failures
                     if f:
                         fail_msg = "%s failed" % (count_noun(f, "doctest"))
-                        if baseline.get('failed', False):
-                            fail_msg += " [failed in baseline]"
-                        postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                        postscript['lines'].append(self.report_head(source, fail_msg))
                         if not baseline.get('failed', False):
                             self.error_status |= 1
                     if f or result_dict.err == 'tab':

From c05bdba849318ae645270685be9085429a0944ba Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 11:23:44 -0800
Subject: [PATCH 15/26] DocTestReporter.report_head: Move all printing of
 'failed in baseline' here

---
 src/sage/doctest/forker.py    |  4 ----
 src/sage/doctest/reporting.py | 37 ++++++++++++++++++++---------------
 2 files changed, 21 insertions(+), 20 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index cb8348e2732..a7379ec97bd 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1725,8 +1725,6 @@ def serial_dispatch(self):
         for source in self.controller.sources:
             heading = self.controller.reporter.report_head(source)
             baseline = self.controller.source_baseline(source)
-            if baseline.get('failed', False):
-                heading += "  # [failed in baseline]"
             if not self.controller.options.only_errors:
                 self.controller.log(heading)
 
@@ -1996,8 +1994,6 @@ def sel_exit():
                                 worker_options.target_walltime = (target_endtime - now) / (max(1, pending_tests / opt.nthreads))
                             w = DocTestWorker(source, options=worker_options, funclist=[sel_exit], baseline=baseline)
                             heading = self.controller.reporter.report_head(w.source)
-                            if baseline.get('failed', False):
-                                heading += "  # [failed in baseline]"
                             if not self.controller.options.only_errors:
                                 w.messages = heading + "\n"
                             # Store length of heading to detect if the
diff --git a/src/sage/doctest/reporting.py b/src/sage/doctest/reporting.py
index dcc0dee792c..42cc3b6d85e 100644
--- a/src/sage/doctest/reporting.py
+++ b/src/sage/doctest/reporting.py
@@ -162,14 +162,16 @@ def were_doctests_with_optional_tag_run(self, tag):
             return True
         return False
 
-    def report_head(self, source):
+    def report_head(self, source, fail_msg=None):
         """
-        Return the "sage -t [options] file.py" line as string.
+        Return the ``sage -t [options] file.py`` line as string.
 
         INPUT:
 
         - ``source`` -- a source from :mod:`sage.doctest.sources`
 
+        - ``fail_msg`` -- ``None`` or a string
+
         EXAMPLES::
 
             sage: from sage.doctest.reporting import DocTestReporter
@@ -190,6 +192,8 @@ def report_head(self, source):
             sage: DD.long = True
             sage: print(DTR.report_head(FDS))
             sage -t --long .../sage/doctest/reporting.py
+            sage: print(DTR.report_head(FDS, "Failed by self-sabotage"))
+            sage -t --long .../sage/doctest/reporting.py  # Failed by self-sabotage
         """
         cmd = "sage -t"
         if self.controller.options.long:
@@ -206,6 +210,13 @@ def report_head(self, source):
         if environment != "sage.repl.ipython_kernel.all_jupyter":
             cmd += f" --environment={environment}"
         cmd += " " + source.printpath
+        baseline = self.controller.source_baseline(source)
+        if fail_msg:
+            cmd += "  # " + fail_msg
+        if baseline.get('failed', False):
+            if not fail_msg:
+                cmd += "  #"
+            cmd += " [failed in baseline]"
         return cmd
 
     def report(self, source, timeout, return_code, results, output, pid=None):
@@ -420,12 +431,10 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                         fail_msg += " (and interrupt failed)"
                     else:
                         fail_msg += " (with %s after interrupt)" % signal_name(sig)
-                if baseline.get('failed', False):
-                    fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s timed out:" % (fail_msg, "*"*70, process_name))
                 log(output)
                 log("*"*70)
-                postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                postscript['lines'].append(self.report_head(source, fail_msg))
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
                 if not baseline.get('failed', False):
                     self.error_status |= 4
@@ -436,12 +445,10 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     fail_msg = "Killed due to %s" % signal_name(-return_code)
                 if ntests > 0:
                     fail_msg += " after testing finished"
-                if baseline.get('failed', False):
-                    fail_msg += " [failed in baseline]"
                 log("    %s\n%s\nTests run before %s failed:" % (fail_msg,"*"*70, process_name))
                 log(output)
                 log("*"*70)
-                postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                postscript['lines'].append(self.report_head(source, fail_msg))
                 stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
                 if not baseline.get('failed', False):
                     self.error_status |= (8 if return_code > 0 else 16)
@@ -458,13 +465,13 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     log("    Error in doctesting framework (bad result returned)\n%s\nTests run before error:" % ("*"*70))
                     log(output)
                     log("*"*70)
-                    postscript['lines'].append(cmd + "  # Testing error: bad result")
+                    postscript['lines'].append(self.report_head(source, "Testing error: bad result"))
                     self.error_status |= 64
                 elif result_dict.err == 'noresult':
                     log("    Error in doctesting framework (no result returned)\n%s\nTests run before error:" % ("*"*70))
                     log(output)
                     log("*"*70)
-                    postscript['lines'].append(cmd + "  # Testing error: no result")
+                    postscript['lines'].append(self.report_head(source, "Testing error: no result"))
                     self.error_status |= 64
                 elif result_dict.err == 'tab':
                     if len(result_dict.tab_linenos) > 5:
@@ -473,11 +480,11 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     if len(result_dict.tab_linenos) > 1:
                         tabs = "s" + tabs
                     log("    Error: TAB character found at line%s" % (tabs))
-                    postscript['lines'].append(cmd + "  # Tab character found")
+                    postscript['lines'].append(self.report_head(source, "Tab character found"))
                     self.error_status |= 32
                 elif result_dict.err == 'line_number':
                     log("    Error: Source line number found")
-                    postscript['lines'].append(cmd + "  # Source line number found")
+                    postscript['lines'].append(self.report_head(source, "Source line number found"))
                     self.error_status |= 256
                 elif result_dict.err is not None:
                     # This case should not occur
@@ -494,7 +501,7 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     if output:
                         log("Tests run before doctest exception:\n" + output)
                         log("*"*70)
-                    postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                    postscript['lines'].append(self.report_head(source, fail_msg))
                     if hasattr(result_dict, 'tb'):
                         log(result_dict.tb)
                     if hasattr(result_dict, 'walltime'):
@@ -506,9 +513,7 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                     f = result_dict.failures
                     if f:
                         fail_msg = "%s failed" % (count_noun(f, "doctest"))
-                        if baseline.get('failed', False):
-                            fail_msg += " [failed in baseline]"
-                        postscript['lines'].append(cmd + "  # %s" % fail_msg)
+                        postscript['lines'].append(self.report_head(source, fail_msg))
                         if not baseline.get('failed', False):
                             self.error_status |= 1
                     if f or result_dict.err == 'tab':

From 179ebb4d4f2519fd069cbccfed240251fb48e41e Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 12:11:15 -0800
Subject: [PATCH 16/26] .github/workflows/ci-conda-known-test-failures.json:
 Add references from commit messages

---
 .../ci-conda-known-test-failures.json         | 28 +++++++++----------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 12ae507bedc..348815e5720 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -3,13 +3,13 @@
         "failed": true
     },
     "sage.algebras.fusion_rings.fusion_ring": {
-        "failed": true
+        "failed": "see https://github.com/sagemath/sage/pull/36372"
     },
     "sage.combinat.cluster_algebra_quiver.quiver": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6836592771/job/18591690058#step:11:10059"
     },
     "sage.geometry.cone": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6827937663/job/18571052628#step:11:12362"
     },
     "sage.groups.matrix_gps.finitely_generated_gap": {
         "failed": true
@@ -18,31 +18,31 @@
         "failed": true
     },
     "sage.libs.gap.element": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6840579851/job/18600012965#step:11:13016"
     },
     "sage.libs.singular.singular": {
         "failed": true
     },
     "sage.matrix.matrix2": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6835143781/job/18588599649#step:11:16939"
     },
     "sage.matrix.matrix_integer_sparse": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6827937663/job/18571052628#step:11:12362"
     },
     "sage.misc.lazy_import": {
         "failed": true
     },
     "sage.misc.weak_dict": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6870325919/job/18684964234#step:11:10059"
     },
     "sage.modular.modform.l_series_gross_zagier": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6859244281/job/18651257775#step:11:10102"
     },
     "sage.parallel.map_reduce": {
-        "failed": true
+        "failed": "random failure https://github.com/sagemath/sage/issues/36939"
     },
     "sage.rings.function_field.drinfeld_modules.morphism": {
-        "failed": true
+        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6840502530/job/18599835766#step:11:10107"
     },
     "sage.rings.polynomial.multi_polynomial_ideal": {
         "failed": true
@@ -51,19 +51,19 @@
         "failed": true
     },
     "sage.rings.polynomial.polynomial_element": {
-        "failed": true
+        "failed": "see https://github.com/sagemath/sage/pull/36372"
     },
     "sage.rings.polynomial.skew_polynomial_finite_field": {
         "failed": true
     },
     "sage.schemes.elliptic_curves.ell_field": {
-        "failed": true
+        "failed": "random failure https://github.com/sagemath/sage/issues/36832"
     },
     "sage.sets.recursively_enumerated_set": {
-        "failed": true
+        "failed": "see https://github.com/sagemath/sage/pull/36372"
     },
     "sage.structure.coerce_actions": {
-        "failed": true
+        "failed": "random failure https://github.com/sagemath/sage/issues/35973"
     },
     "sage.tests.gap_packages": {
         "failed": true

From b41999bf330d591480cbe88d0b4893feb63a26a4 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 12:40:00 -0800
Subject: [PATCH 17/26] .github/workflows/ci-conda-known-test-failures.json:
 Add details from https://github.com/sagemath/sage/pull/36372

---
 .../ci-conda-known-test-failures.json         | 20 ++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/ci-conda-known-test-failures.json b/.github/workflows/ci-conda-known-test-failures.json
index 348815e5720..47d0adcd8e9 100644
--- a/.github/workflows/ci-conda-known-test-failures.json
+++ b/.github/workflows/ci-conda-known-test-failures.json
@@ -1,15 +1,18 @@
 {
+    "doc.en.constructions.calculus": {
+        "failed": "unreported random failures in plotting"
+    },
     "sage_setup.clean": {
-        "failed": true
+        "failed": "_find_stale_files finds some stale files under sage/tests when executed under conda"
     },
     "sage.algebras.fusion_rings.fusion_ring": {
-        "failed": "see https://github.com/sagemath/sage/pull/36372"
+        "failed": "unreported random timeouts"
     },
     "sage.combinat.cluster_algebra_quiver.quiver": {
         "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6836592771/job/18591690058#step:11:10059"
     },
     "sage.geometry.cone": {
-        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6827937663/job/18571052628#step:11:12362"
+        "failed": "unreported random timeouts seen in https://github.com/sagemath/sage/actions/runs/6827937663/job/18571052628#step:11:12362"
     },
     "sage.groups.matrix_gps.finitely_generated_gap": {
         "failed": true
@@ -36,13 +39,16 @@
         "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6870325919/job/18684964234#step:11:10059"
     },
     "sage.modular.modform.l_series_gross_zagier": {
-        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6859244281/job/18651257775#step:11:10102"
+        "failed": "unreported failure seen in https://github.com/sagemath/sage/actions/runs/6859244281/job/18651257775#step:11:10102"
     },
     "sage.parallel.map_reduce": {
         "failed": "random failure https://github.com/sagemath/sage/issues/36939"
     },
+    "sage.plot.plot": {
+        "failed": "unreported random failure (macOS)"
+    },
     "sage.rings.function_field.drinfeld_modules.morphism": {
-        "failed": "failure seen in https://github.com/sagemath/sage/actions/runs/6840502530/job/18599835766#step:11:10107"
+        "failed": "unreported random failure seen in https://github.com/sagemath/sage/actions/runs/6840502530/job/18599835766#step:11:10107"
     },
     "sage.rings.polynomial.multi_polynomial_ideal": {
         "failed": true
@@ -51,7 +57,7 @@
         "failed": true
     },
     "sage.rings.polynomial.polynomial_element": {
-        "failed": "see https://github.com/sagemath/sage/pull/36372"
+        "failed": "unreported random failure in symbolic 'roots' (macOS)"
     },
     "sage.rings.polynomial.skew_polynomial_finite_field": {
         "failed": true
@@ -60,7 +66,7 @@
         "failed": "random failure https://github.com/sagemath/sage/issues/36832"
     },
     "sage.sets.recursively_enumerated_set": {
-        "failed": "see https://github.com/sagemath/sage/pull/36372"
+        "failed": "random failures related to AlarmInterrupt (macOS)"
     },
     "sage.structure.coerce_actions": {
         "failed": "random failure https://github.com/sagemath/sage/issues/35973"

From 63b8b108fee7e18a8c51d7e1460cea7ad9a02804 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 21 Dec 2023 12:40:58 -0800
Subject: [PATCH 18/26] src/sage/doctest/reporting.py: Include message in
 '[failed in baseline]' if provided

---
 src/sage/doctest/reporting.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/src/sage/doctest/reporting.py b/src/sage/doctest/reporting.py
index 42cc3b6d85e..a1e85af1a57 100644
--- a/src/sage/doctest/reporting.py
+++ b/src/sage/doctest/reporting.py
@@ -213,10 +213,13 @@ def report_head(self, source, fail_msg=None):
         baseline = self.controller.source_baseline(source)
         if fail_msg:
             cmd += "  # " + fail_msg
-        if baseline.get('failed', False):
+        if failed := baseline.get('failed', False):
             if not fail_msg:
                 cmd += "  #"
-            cmd += " [failed in baseline]"
+            if failed is True:
+                cmd += " [failed in baseline]"
+            else:
+                cmd += f" [failed in baseline: {failed}]"
         return cmd
 
     def report(self, source, timeout, return_code, results, output, pid=None):

From 4b0f525f6d24fc1167dd1b9375fb3d920d7fbe26 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Fri, 22 Dec 2023 20:00:06 -0800
Subject: [PATCH 19/26] .github/workflows/build.yml: Fix up path to coverage
 files

---
 .github/workflows/build.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 2b87d6edb60..2f5e20f2dff 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -192,7 +192,7 @@ jobs:
       - name: Prepare coverage results
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
-          ./venv/bin/python3 -m coverage combine src/.coverage/
+          ./venv/bin/python3 -m coverage combine .coverage/
           ./venv/bin/python3 -m coverage xml
           mkdir -p coverage-report
           mv coverage.xml coverage-report/

From 0bfd2c933d671cb01f14733ba6eedd8bb149e498 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Fri, 22 Dec 2023 21:31:22 -0800
Subject: [PATCH 20/26] src/sage/doctest/reporting.py: Make 'AlarmInterrupt in
 doctesting framework' baseline-aware

---
 src/sage/doctest/reporting.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/sage/doctest/reporting.py b/src/sage/doctest/reporting.py
index 42cc3b6d85e..cda0e8a3f20 100644
--- a/src/sage/doctest/reporting.py
+++ b/src/sage/doctest/reporting.py
@@ -508,7 +508,8 @@ def report(self, source, timeout, return_code, results, output, pid=None):
                         stats[basename] = {"failed": True, "walltime": wall, "ntests": ntests}
                     else:
                         stats[basename] = {"failed": True, "walltime": 1e6, "ntests": ntests}
-                    self.error_status |= 64
+                    if not baseline.get('failed', False):  # e.g. AlarmInterrupt in doctesting framework
+                        self.error_status |= 64
                 if result_dict.err is None or result_dict.err == 'tab':
                     f = result_dict.failures
                     if f:

From ce3b849dc841e269357ef260ec6a9e1088143cc1 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Wed, 27 Dec 2023 20:18:36 -0800
Subject: [PATCH 21/26] .github/workflows/build.yml: Do not create GitHub
 Annotations for the tests of the new files, to avoid duplication

---
 .github/workflows/build.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 2f5e20f2dff..71ba6d15351 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -158,7 +158,7 @@ jobs:
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
           # We run tests with "sage -t --new"; this only tests the uncommitted changes.
-          ./sage -t --new -p2 --format github
+          ./sage -t --new -p2
         working-directory: ./worktree-image
         env:
           MAKE: make -j2 --output-sync=recurse

From b83dcf3102dd1b8a6d5c869895c4a80c9379e9aa Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Fri, 29 Dec 2023 08:38:30 -0800
Subject: [PATCH 22/26] src/sage/doctest/forker.py: Use urllib.parse.quote for
 github annotations

---
 src/sage/doctest/forker.py | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index 88887240c9f..e972031fbd1 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1248,16 +1248,21 @@ def _failure_header(self, test, example, message='Failed example:'):
                 out.append(message)
             elif self.options.format == 'github':
                 # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
+                from urllib.parse import quote
+
+                def q(string):
+                    return quote(string, safe=' /,"\'')
+
                 if message.startswith('Warning: '):
-                    command = f'::warning title={message}'
+                    command = f'::warning title={q(message)}'
                     message = message[len('Warning: '):]
                 elif self.baseline.get('failed', False):
-                    command = f'::notice title={message}'
+                    command = f'::notice title={q(message)}'
                     message += ' [failed in baseline]'
                 else:
-                    command = f'::error title={message}'
+                    command = f'::error title={q(message)}'
                 if extra := getattr(example, 'extra', None):
-                    message += f': {extra}'
+                    message += f': {q(extra)}'
                 if test.filename:
                     command += f',file={test.filename}'
                     if test.lineno is not None and example.lineno is not None:
@@ -1266,7 +1271,7 @@ def _failure_header(self, test, example, message='Failed example:'):
                     lineno = None
                 else:
                     command += f',line={example.lineno + 1}'
-                command += f'::{message}'
+                command += f'::{q(message)}'
                 out.append(command)
             else:
                 raise ValueError(f'unknown format option: {self.options.format}')
@@ -1442,7 +1447,7 @@ def report_failure(self, out, test, example, got, globs):
         """
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
-            example.extra = f'Got: {got!r}'
+            example.extra = f'Got: {got}'
             returnval = doctest.DocTestRunner.report_failure(self, out, test, example, got)
             if self.options.debug:
                 self._fakeout.stop_spoofing()
@@ -1594,7 +1599,7 @@ def report_unexpected_exception(self, out, test, example, exc_info):
         if not self.options.initial or self.no_failure_yet:
             self.no_failure_yet = False
 
-            example.extra = "Exception raised: " + repr("".join(traceback.format_exception(*exc_info)))
+            example.extra = "Exception raised:\n" + "".join(traceback.format_exception(*exc_info))
             returnval = doctest.DocTestRunner.report_unexpected_exception(self, out, test, example, exc_info)
             if self.options.debug:
                 self._fakeout.stop_spoofing()

From 96e043138e420c26c76171d77073e05b5855c26b Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Fri, 29 Dec 2023 11:43:39 -0800
Subject: [PATCH 23/26] src/sage/doctest/forker.py: Don't use
 urllib.parse.quote, only quote \n, pad short multi-line messages

---
 src/sage/doctest/forker.py | 31 +++++++++++++++++++++----------
 1 file changed, 21 insertions(+), 10 deletions(-)

diff --git a/src/sage/doctest/forker.py b/src/sage/doctest/forker.py
index e972031fbd1..67e587c2c9d 100644
--- a/src/sage/doctest/forker.py
+++ b/src/sage/doctest/forker.py
@@ -1248,21 +1248,16 @@ def _failure_header(self, test, example, message='Failed example:'):
                 out.append(message)
             elif self.options.format == 'github':
                 # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#using-workflow-commands-to-access-toolkit-functions
-                from urllib.parse import quote
-
-                def q(string):
-                    return quote(string, safe=' /,"\'')
-
                 if message.startswith('Warning: '):
-                    command = f'::warning title={q(message)}'
+                    command = f'::warning title={message}'
                     message = message[len('Warning: '):]
                 elif self.baseline.get('failed', False):
-                    command = f'::notice title={q(message)}'
+                    command = f'::notice title={message}'
                     message += ' [failed in baseline]'
                 else:
-                    command = f'::error title={q(message)}'
+                    command = f'::error title={message}'
                 if extra := getattr(example, 'extra', None):
-                    message += f': {q(extra)}'
+                    message += f': {extra}'
                 if test.filename:
                     command += f',file={test.filename}'
                     if test.lineno is not None and example.lineno is not None:
@@ -1271,7 +1266,23 @@ def q(string):
                     lineno = None
                 else:
                     command += f',line={example.lineno + 1}'
-                command += f'::{q(message)}'
+                #
+                # Urlencoding trick for multi-line annotations
+                # https://github.com/actions/starter-workflows/issues/68#issuecomment-581479448
+                #
+                # This only affects the display in the workflow Summary, after clicking "Show more";
+                # the message needs to be long enough so that "Show more" becomes available.
+                # https://github.com/actions/toolkit/issues/193#issuecomment-1867084340
+                #
+                # Unfortunately, this trick does not make the annotations in the diff view multi-line.
+                #
+                if '\n' in message:
+                    message = message.replace('\n', '%0A')
+                    # The actual threshold for "Show more" to appear depends on the window size.
+                    show_more_threshold = 500
+                    if (pad := show_more_threshold - len(message)) > 0:
+                        message += ' ' * pad
+                command += f'::{message}'
                 out.append(command)
             else:
                 raise ValueError(f'unknown format option: {self.options.format}')

From 1d69f0798f503eb457ff44e2c7f0c96b5acf3495 Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Fri, 29 Dec 2023 14:57:58 -0800
Subject: [PATCH 24/26] .github/workflows/build.yml: Fix "Couldn't combine from
 non-existent path '.coverage/'"

---
 .github/workflows/build.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 71ba6d15351..6af89c85d71 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -192,7 +192,7 @@ jobs:
       - name: Prepare coverage results
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
-          ./venv/bin/python3 -m coverage combine .coverage/
+          ./venv/bin/python3 -m coverage combine
           ./venv/bin/python3 -m coverage xml
           mkdir -p coverage-report
           mv coverage.xml coverage-report/

From 173bc417bdec2b02b19e72c7a95a7dcbbd41a86a Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 4 Jan 2024 16:04:44 -0800
Subject: [PATCH 25/26] .github/workflows/build.yml: Pass --rcfile to coverage

---
 .github/workflows/build.yml | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 6af89c85d71..94fa81e4e8b 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -175,9 +175,9 @@ jobs:
       - name: Pytest
         if: contains(github.ref, 'pytest')
         run: |
-          ../sage -python -m pip install coverage pytest-xdist
-          ../sage -python -m coverage run -m pytest -c tox.ini --doctest-modules || true
-        working-directory: ./worktree-image/src
+          ./sage -python -m pip install coverage pytest-xdist
+          ./sage -python -m coverage run --rcfile=src/tox.ini -m pytest -c src/tox.ini --doctest-modules || true
+        working-directory: ./worktree-image
         env:
           # Increase the length of the lines in the "short summary"
           COLUMNS: 120
@@ -186,14 +186,14 @@ jobs:
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
           ./sage -python -m pip install coverage
-          ./sage -python -m coverage run ./src/bin/sage-runtests --all --long -p2 --format github --random-seed=286735480429121101562228604801325644303
+          ./sage -python -m coverage run --rcfile=src/tox.ini src/bin/sage-runtests --all --long -p2 --format github --random-seed=286735480429121101562228604801325644303
         working-directory: ./worktree-image
 
       - name: Prepare coverage results
         if: (success() || failure()) && steps.build.outcome == 'success'
         run: |
-          ./venv/bin/python3 -m coverage combine
-          ./venv/bin/python3 -m coverage xml
+          ./sage -python -m coverage combine --rcfile=src/tox.ini
+          ./sage -python -m coverage xml --rcfile=src/tox.ini
           mkdir -p coverage-report
           mv coverage.xml coverage-report/
         working-directory: ./worktree-image

From 4ecf7cad57fd0018cc917b5cc4859871d7e1c2bf Mon Sep 17 00:00:00 2001
From: Matthias Koeppe <mkoeppe@math.ucdavis.edu>
Date: Thu, 4 Jan 2024 19:11:37 -0800
Subject: [PATCH 26/26] src/tox.ini (coverage:report): Ignore errors, ignore
 empty

---
 src/tox.ini | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/src/tox.ini b/src/tox.ini
index 00f7a153f35..f2f1fc158fd 100644
--- a/src/tox.ini
+++ b/src/tox.ini
@@ -299,3 +299,7 @@ source = sage
 concurrency = multiprocessing
 data_file = .coverage/.coverage
 disable_warnings = no-data-collected
+
+[coverage:report]
+ignore_errors = True
+skip_empty = True