dvc-utils 0.0.9__py3-none-any.whl → 0.1.0__py3-none-any.whl

dvc_utils/cli.py

@@ -4,7 +4,8 @@ from typing import Tuple
 
 import click
 from click import option, argument, group
-from utz import diff_cmds, process, err
+from utz import process, err
+from qmdx import join_pipelines
 
 from dvc_utils.path import dvc_paths, dvc_path as dvc_cache_path
 
@@ -62,39 +63,32 @@ def dvc_utils_diff(
         raise ValueError(f"Invalid refspec: {refspec}")
 
     log = err if verbose else False
-    before_path = dvc_cache_path(before, dvc_path, log=log)
-    after_path = path if after is None else dvc_cache_path(after, dvc_path, log=log)
-
+    path1 = dvc_cache_path(before, dvc_path, log=log)
+    path2 = path if after is None else dvc_cache_path(after, dvc_path, log=log)
+
+    diff_args = [
+        *(['-w'] if ignore_whitespace else []),
+        *(['-U', str(unified)] if unified is not None else []),
+        *(['--color=always'] if color else []),
+    ]
     if cmds:
         cmd, *sub_cmds = cmds
+        cmds1 = [ f'{cmd} {path1}', *sub_cmds ]
+        cmds2 = [ f'{cmd} {path2}', *sub_cmds ]
         if not shell:
-            sub_cmds = [ shlex.split(c) for c in sub_cmds ]
-            before_cmds = [
-                shlex.split(f'{cmd} {before_path}'),
-                *sub_cmds,
-            ]
-            after_cmds = [
-                shlex.split(f'{cmd} {after_path}'),
-                *sub_cmds,
-            ]
-            shell_kwargs = {}
-        else:
-            before_cmds = [ f'{cmd} {before_path}', *sub_cmds ]
-            after_cmds = [ f'{cmd} {after_path}', *sub_cmds ]
-            shell_kwargs = dict(shell=shell)
-
-        diff_cmds(
-            before_cmds,
-            after_cmds,
+            cmds1 = [ shlex.split(cmd) for cmd in cmds1 ]
+            cmds2 = [ shlex.split(cmd) for cmd in cmds2 ]
+
+        join_pipelines(
+            base_cmd=['diff', *diff_args],
+            cmds1=cmds1,
+            cmds2=cmds2,
             verbose=verbose,
-            color=color,
-            unified=unified,
-            ignore_whitespace=ignore_whitespace,
+            shell=shell,
             shell_executable=shell_executable,
-            **shell_kwargs,
         )
     else:
-        process.run('diff', before_path, after_path, log=log)
+        process.run('diff', *diff_args, path1, path2, log=log)
 
 
 if __name__ == '__main__':

{dvc_utils-0.0.9.dist-info → dvc_utils-0.1.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dvc-utils
-Version: 0.0.9
+Version: 0.1.0
 Summary: CLI for diffing DVC files at two commits (or one commit vs. current worktree), optionally passing both through another command first
 Home-page: https://github.com/runsascoded/dvc-utils
 Author: Ryan Williams
@@ -8,10 +8,16 @@ Author-email: ryan@runsascoded.com
 License: MIT
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: click
+Requires-Dist: pyyaml
+Requires-Dist: qmdx
+Requires-Dist: utz>=0.11.3
 
 # dvc-utils
 Diff [DVC] files, optionally piping through other commands first.
 
+[![dvc-utils on PyPI](https://img.shields.io/pypi/v/dvc-utils?label=dvc-utils)][PyPI]
+
 <!-- toc -->
 - [Installation](#installation)
 - [Usage](#usage)
@@ -246,7 +252,7 @@ pel "ddcr $r guc h1 spc kq kcr snc 'sdf seds' sort"
 
 <details>
 <summary>
-Aliases used in the pipeline:
+Explanation of aliases
 </summary>
 
 - [`gdno`] (`git diff --name-only`): list files changed in the given commit range and directory
@@ -254,9 +260,9 @@ Aliases used in the pipeline:
 - [`ddcr`] (`dvc-diff -cr`): colorized `diff` output, revision range `$r`
 - [`guc`] (`gunzip -c`): uncompress the `.csv.gz` files
 - [`h1`] (`head -n1`): only examine each file's header line
-- [`spc`] (`tr , $'\n'`): split the header line by commas (so each column name will be on one line, for easier `diff`ing below)
-- [`kq`] (`tr -d '"'`): kill quote characters (in this case, header-column name quoting changed, but I don't care about that)
-- [`kcr`] (`tr -d '\r'`): kill carriage returns (line endings also changed)
+- [`spc`] (`tr , $'\n'`): **sp**lit the header line by **c**ommas (so each column name will be on one line, for easier `diff`ing below)
+- [`kq`] (`tr -d '"'`): **k**ill **q**uote characters (in this case, header-column name quoting changed, but I don't care about that)
+- [`kcr`] (`tr -d '\r'`): **k**ill **c**arriage **r**eturns (line endings also changed)
 - [`snc`] (`sed -f 'snake_case.sed'`): snake-case column names
 - [`sdf`] (`sed -f`): execute the `sed` substitution commands defined in the `seds` file above
 - `sort`: sort the column names alphabetically (to identify missing or added columns, ignore rearrangements)
@@ -311,6 +317,7 @@ s3/ctbk/csvs/202003-citibike-tripdata.csv.gz.dvc:
 This helped me see that the data update in question (`c0..c1`) dropped some fields (`bikeid, birth_year`, `gender`, `tripduration`) and added others (`ride_id`, `rideable_type`), for `202001` and later.
 
 [DVC]: https://dvc.org/
+[PyPI]: https://pypi.org/project/dvc-utils/
 [`parquet2json`]: https://github.com/jupiter/parquet2json
 [hudcostreets/nj-crashes]: https://github.com/hudcostreets/nj-crashes
 [Parquet]: https://parquet.apache.org/

dvc_utils-0.1.0.dist-info/RECORD

@@ -0,0 +1,9 @@
+dvc_utils/__init__.py,sha256=mP-p1Sl2JMMShM_hRhu86pFNfIq_8E_feh1CN47LWcs,86
+dvc_utils/cli.py,sha256=CcOa5Qmkry3PASz8nu_XqcBB6GnGmcf0e8zsuqSDsCM,3534
+dvc_utils/path.py,sha256=PoAbeaqRPDksY2hcUeF8xZ6Nr6hLIZprey3VNT4V5bc,1727
+dvc_utils-0.1.0.dist-info/LICENSE,sha256=ZS8AReay7xmQzBAHwxIuTouGXz3SKgUa2_Sz8Ip0EzQ,1070
+dvc_utils-0.1.0.dist-info/METADATA,sha256=F3GUp8NMg0oEqsqtI_jECiWkoO7-sGYb8KJT4upGqNM,11722
+dvc_utils-0.1.0.dist-info/WHEEL,sha256=bFJAMchF8aTQGUgMZzHJyDDMPTO3ToJ7x23SLJa1SVo,92
+dvc_utils-0.1.0.dist-info/entry_points.txt,sha256=luxCQr8OS-jMSyyDhB9KDQhUbP8UH6UMcy-vkfXX7Gg,88
+dvc_utils-0.1.0.dist-info/top_level.txt,sha256=jT0-PJa2t_eFRE9rn-52AjdnZ8nQeEHllf2kJmaGh80,10
+dvc_utils-0.1.0.dist-info/RECORD,,

dvc_utils-0.0.9.dist-info/RECORD

@@ -1,9 +0,0 @@
-dvc_utils/__init__.py,sha256=mP-p1Sl2JMMShM_hRhu86pFNfIq_8E_feh1CN47LWcs,86
-dvc_utils/cli.py,sha256=D5GGDPpMlKG-8IHkTXPIB2z-QAKHzbm5gQXqj7vCquY,3701
-dvc_utils/path.py,sha256=PoAbeaqRPDksY2hcUeF8xZ6Nr6hLIZprey3VNT4V5bc,1727
-dvc_utils-0.0.9.dist-info/LICENSE,sha256=ZS8AReay7xmQzBAHwxIuTouGXz3SKgUa2_Sz8Ip0EzQ,1070
-dvc_utils-0.0.9.dist-info/METADATA,sha256=zhFkhQbnbOYwwhHJiXUj8otqRw5tfmVsXx-b5MlByvw,11480
-dvc_utils-0.0.9.dist-info/WHEEL,sha256=bFJAMchF8aTQGUgMZzHJyDDMPTO3ToJ7x23SLJa1SVo,92
-dvc_utils-0.0.9.dist-info/entry_points.txt,sha256=luxCQr8OS-jMSyyDhB9KDQhUbP8UH6UMcy-vkfXX7Gg,88
-dvc_utils-0.0.9.dist-info/top_level.txt,sha256=jT0-PJa2t_eFRE9rn-52AjdnZ8nQeEHllf2kJmaGh80,10
-dvc_utils-0.0.9.dist-info/RECORD,,