tag: getting rid of it (#3699)

* tags: get rid of it

* tests: adjust for tags and allow @ in stage name

* disallow @ on stage name

* fix ds issue, leftover tag in Stage

Author: skshetry

dvc/cli.py

@@ -31,7 +31,6 @@
     repro,
     root,
     run,
-    tag,
     unprotect,
     update,
     version,
@@ -71,7 +70,6 @@
     pipeline,
     daemon,
     commit,
-    tag,
     diff,
     version,
     update,

dvc/command/pipeline.py

@@ -14,8 +14,8 @@ def _show(self, target, commands, outs, locked):
         from dvc import dvcfile
         from dvc.utils import parse_target
 
-        path, name, tag = parse_target(target)
-        stage = dvcfile.Dvcfile(self.repo, path, tag=tag).stages[name]
+        path, name = parse_target(target)
+        stage = dvcfile.Dvcfile(self.repo, path).stages[name]
         G = self.repo.graph
         stages = networkx.dfs_postorder_nodes(G, stage)
         if locked:
@@ -38,8 +38,8 @@ def _build_graph(self, target, commands=False, outs=False):
         from dvc.repo.graph import get_pipeline
         from dvc.utils import parse_target
 
-        path, name, tag = parse_target(target)
-        target_stage = dvcfile.Dvcfile(self.repo, path, tag=tag).stages[name]
+        path, name = parse_target(target)
+        target_stage = dvcfile.Dvcfile(self.repo, path).stages[name]
         G = get_pipeline(self.repo.pipelines, target_stage)
 
         nodes = set()

dvc/command/tag.py

@@ -128,19 +128,18 @@ def dump(self, stage, **kwargs):
 class SingleStageFile(FileMixin):
     from dvc.schema import COMPILED_SINGLE_STAGE_SCHEMA as SCHEMA
 
-    def __init__(self, repo, path, tag=None):
+    def __init__(self, repo, path):
         super().__init__(repo, path)
-        self.tag = tag
 
     @property
     def stage(self):
         data, raw = self._load()
-        return SingleStageLoader.load_stage(self, data, raw, tag=self.tag)
+        return SingleStageLoader.load_stage(self, data, raw)
 
     @property
     def stages(self):
         data, raw = self._load()
-        return SingleStageLoader(self, data, raw, tag=self.tag)
+        return SingleStageLoader(self, data, raw)
 
     def dump(self, stage, **kwargs):
         """Dumps given stage appropriately in the dvcfile."""

dvc/dvcfile.py

@@ -49,17 +49,14 @@
     HDFSRemote.PARAM_CHECKSUM: CHECKSUM_SCHEMA,
 }
 
-TAGS_SCHEMA = {str: CHECKSUMS_SCHEMA}
-
 SCHEMA = CHECKSUMS_SCHEMA.copy()
 SCHEMA[Required(BaseOutput.PARAM_PATH)] = str
 SCHEMA[BaseOutput.PARAM_CACHE] = bool
 SCHEMA[BaseOutput.PARAM_METRIC] = BaseOutput.METRIC_SCHEMA
-SCHEMA[BaseOutput.PARAM_TAGS] = TAGS_SCHEMA
 SCHEMA[BaseOutput.PARAM_PERSIST] = bool
 
 
-def _get(stage, p, info, cache, metric, persist=False, tags=None):
+def _get(stage, p, info, cache, metric, persist=False):
     parsed = urlparse(p)
 
     if parsed.scheme == "remote":
@@ -72,7 +69,6 @@ def _get(stage, p, info, cache, metric, persist=False, tags=None):
             remote=remote,
             metric=metric,
             persist=persist,
-            tags=tags,
         )
 
     for o in OUTS:
@@ -85,7 +81,6 @@ def _get(stage, p, info, cache, metric, persist=False, tags=None):
                 remote=None,
                 metric=metric,
                 persist=persist,
-                tags=tags,
             )
     return LocalOutput(
         stage,
@@ -95,7 +90,6 @@ def _get(stage, p, info, cache, metric, persist=False, tags=None):
         remote=None,
         metric=metric,
         persist=persist,
-        tags=tags,
     )
 
 
@@ -106,16 +100,9 @@ def loadd_from(stage, d_list):
         cache = d.pop(BaseOutput.PARAM_CACHE, True)
         metric = d.pop(BaseOutput.PARAM_METRIC, False)
         persist = d.pop(BaseOutput.PARAM_PERSIST, False)
-        tags = d.pop(BaseOutput.PARAM_TAGS, None)
         ret.append(
             _get(
-                stage,
-                p,
-                info=d,
-                cache=cache,
-                metric=metric,
-                persist=persist,
-                tags=tags,
+                stage, p, info=d, cache=cache, metric=metric, persist=persist,
             )
         )
     return ret

dvc/output/__init__.py

@@ -59,8 +59,6 @@ class BaseOutput(object):
         },
     )
 
-    PARAM_TAGS = "tags"
-
     DoesNotExistError = OutputDoesNotExistError
     IsNotFileOrDirError = OutputIsNotFileOrDirError
     IsStageFileError = OutputIsStageFileError
@@ -76,7 +74,6 @@ def __init__(
         cache=True,
         metric=False,
         persist=False,
-        tags=None,
     ):
         self._validate_output_path(path)
         # This output (and dependency) objects have too many paths/urls
@@ -97,7 +94,6 @@ def __init__(
         self.use_cache = False if self.IS_DEPENDENCY else cache
         self.metric = False if self.IS_DEPENDENCY else metric
         self.persist = persist
-        self.tags = None if self.IS_DEPENDENCY else (tags or {})
 
         self.path_info = self._parse_path(remote, path)
         if self.use_cache and self.cache is None:
@@ -274,9 +270,6 @@ def dumpd(self):
         ret[self.PARAM_METRIC] = self.metric
         ret[self.PARAM_PERSIST] = self.persist
 
-        if self.tags:
-            ret[self.PARAM_TAGS] = self.tags
-
         return ret
 
     def verify_metric(self):
@@ -291,7 +284,6 @@ def checkout(
         self,
         force=False,
         progress_callback=None,
-        tag=None,
         relink=False,
         filter_info=None,
     ):
@@ -302,14 +294,9 @@ def checkout(
                 )
             return None
 
-        if tag:
-            info = self.tags[tag]
-        else:
-            info = self.info
-
         return self.cache.checkout(
             self.path_info,
-            info,
+            self.info,
             force=force,
             progress_callback=progress_callback,
             relink=relink,

dvc/output/base.py

@@ -1,3 +1,4 @@
+import logging
 import os
 from contextlib import contextmanager
 from functools import wraps
@@ -70,7 +71,6 @@ def __init__(self, root_dir=None):
         from dvc.repo.metrics import Metrics
         from dvc.repo.params import Params
         from dvc.scm.tree import WorkingTree
-        from dvc.repo.tag import Tag
         from dvc.utils.fs import makedirs
 
         root_dir = self.find_root(root_dir)
@@ -106,7 +106,6 @@ def __init__(self, root_dir=None):
 
         self.metrics = Metrics(self)
         self.params = Params(self)
-        self.tag = Tag(self)
 
         self._ignore()
 
@@ -211,8 +210,8 @@ def collect(self, target, with_deps=False, recursive=False, graph=None):
                 os.path.abspath(target), graph or self.graph
             )
 
-        file, name, tag = parse_target(target)
-        dvcfile = Dvcfile(self, file, tag=tag)
+        path, name = parse_target(target)
+        dvcfile = Dvcfile(self, path)
         stages = list(dvcfile.stages.filter(name).values())
         if not with_deps:
             return stages
@@ -229,10 +228,10 @@ def collect_granular(self, target, *args, **kwargs):
         if not target:
             return [(stage, None) for stage in self.stages]
 
-        file, name, tag = parse_target(target)
+        file, name = parse_target(target)
         if is_valid_filename(file) and not kwargs.get("with_deps"):
             # Optimization: do not collect the graph for a specific .dvc target
-            stages = Dvcfile(self, file, tag=tag).stages.filter(name)
+            stages = Dvcfile(self, file).stages.filter(name)
             return [(stage, None) for stage in stages.values()]
 
         try:
@@ -433,7 +432,9 @@ def _collect_stages(self):
         for root, dirs, files in self.tree.walk(self.root_dir):
             for file_name in filter(is_valid_filename, files):
                 path = os.path.join(root, file_name)
-                stages.extend(list(Dvcfile(self, path).stages.values()))
+                stage_loader = Dvcfile(self, path).stages
+                with stage_loader.log_level(at=logging.DEBUG):
+                    stages.extend(stage_loader.values())
                 outs.update(
                     out.fspath
                     for stage in stages