diff: reimplement interface and tests from scratch

dvc/command/diff.py

@@ -1,12 +1,11 @@
 import argparse
+import json
 import logging
+import os
 
-import humanize
-import inflect
-from funcy import compact
+import colorama
 
-from dvc.command.base import append_doc_link
-from dvc.command.base import CmdBase
+from dvc.command.base import CmdBase, append_doc_link
 from dvc.exceptions import DvcException
 
 
@@ -15,156 +14,150 @@
 
 class CmdDiff(CmdBase):
     @staticmethod
-    def _print_size(size):
-        if size < 0:
-            change = "decreased by {}"
-        elif size > 0:
-            change = "increased by {}"
-        else:
-            change = "not changed"
-        natur_size = humanize.naturalsize(abs(size))
-        return change.format(natur_size)
+    def _format(diff):
+        """
+        Given a diff structure, generate a string of paths separated
+        by new lines and grouped together by their state.
+
+        A group's header is colored and its entries are sorted to enhance
+        readability, for example:
+
+            Added:
+                another_file.txt
+                backup.tar
+                dir/
+                dir/1
+
+        An example of a diff formatted when entries contain checksums:
+
+            Added:
+                d3b07384 foo
+
+            Modified:
+                c157a790..f98bf6f1 bar
+
+        If a group has no entries, it won't be included in the result.
+
+        At the bottom, include a summary with the number of files per state.
+        """
+
+        def _digest(checksum):
+            if type(checksum) is str:
+                return checksum[0:8]
+            return "{}..{}".format(checksum["old"][0:8], checksum["new"][0:8])
+
+        colors = {
+            "added": colorama.Fore.GREEN,
+            "modified": colorama.Fore.YELLOW,
+            "deleted": colorama.Fore.RED,
+        }
+
+        summary = {}
+        groups = []
+
+        for state in ["added", "deleted", "modified"]:
+            summary[state] = 0
+            entries = diff[state]
+
+            if not entries:
+                continue
+
+            content = []
+
+            for entry in entries:
+                path = entry["path"]
+                checksum = entry.get("checksum")
+                summary[state] += 1 if not path.endswith(os.sep) else 0
+                content.append(
+                    "{space}{checksum}{separator}{path}".format(
+                        space="    ",
+                        checksum=_digest(checksum) if checksum else "",
+                        separator="  " if checksum else "",
+                        path=entry["path"],
+                    )
+                )
+
+            groups.append(
+                "{color}{header}{nc}:\n{content}".format(
+                    color=colors[state],
+                    header=state.capitalize(),
+                    nc=colorama.Fore.RESET,
+                    content="\n".join(content),
+                )
+            )
 
-    @staticmethod
-    def _get_md5_string(sign, file_name, checksum):
-        sample_msg = ""
-        if file_name:
-            sample_msg = "{}{} with md5 {}\n"
-            sample_msg = sample_msg.format(sign, file_name, checksum)
-        return sample_msg
-
-    @classmethod
-    def _get_dir_changes(cls, dct):
-        import dvc.repo.diff as diff
-
-        engine = inflect.engine()
-        changes_msg = (
-            "{} {} untouched, {} {} modified, {} {} added, "
-            "{} {} deleted, size was {}"
+        groups.append(
+            "summary: added ({added}), deleted ({deleted}),"
+            " modified ({modified})".format_map(summary)
         )
-        changes_msg = changes_msg.format(
-            dct[diff.DIFF_IDENT],
-            engine.plural("file", dct[diff.DIFF_IDENT]),
-            dct[diff.DIFF_CHANGE],
-            engine.plural("file", dct[diff.DIFF_CHANGE]),
-            dct[diff.DIFF_NEW],
-            engine.plural("file", dct[diff.DIFF_NEW]),
-            dct[diff.DIFF_DEL],
-            engine.plural("file", dct[diff.DIFF_DEL]),
-            cls._print_size(dct[diff.DIFF_SIZE]),
-        )
-        return changes_msg
-
-    @classmethod
-    def _get_file_changes(cls, dct):
-        import dvc.repo.diff as diff
-
-        if (
-            dct.get(diff.DIFF_OLD_FILE)
-            and dct.get(diff.DIFF_NEW_FILE)
-            and dct[diff.DIFF_SIZE] == 0
-        ):
-            msg = "file size was not changed"
-        elif dct.get(diff.DIFF_NEW_FILE):
-            msg = "added file with size {}".format(
-                humanize.naturalsize(dct[diff.DIFF_SIZE])
-            )
-        elif dct.get(diff.DIFF_OLD_FILE):
-            msg = "deleted file with size {}".format(
-                humanize.naturalsize(abs(dct[diff.DIFF_SIZE]))
-            )
-        else:
-            msg = "file was modified, file size {}".format(
-                cls._print_size(dct[diff.DIFF_SIZE])
-            )
-        return msg
 
-    @classmethod
-    def _get_royal_changes(cls, dct):
-        import dvc.repo.diff as diff
+        return "\n\n".join(groups)
 
-        if dct[diff.DIFF_SIZE] != diff.DIFF_SIZE_UNKNOWN:
-            if dct.get("is_dir"):
-                return cls._get_dir_changes(dct)
-            else:
-                return cls._get_file_changes(dct)
-        return "size is ?"
+    def run(self):
+        try:
+            diff = self.repo.diff(self.args.a_ref, self.args.b_ref)
 
-    @classmethod
-    def _show(cls, diff_dct):
-        import dvc.repo.diff as diff
+            if not any(diff.values()):
+                return 0
 
-        msg = "dvc diff from {} to {}".format(
-            diff_dct[diff.DIFF_A_REF], diff_dct[diff.DIFF_B_REF]
-        )
-        if diff_dct.get(diff.DIFF_EQUAL):
-            logger.info(msg)
-            return
-        for dct in diff_dct[diff.DIFF_LIST]:
-            msg += "\n\ndiff for '{}'\n".format(dct[diff.DIFF_TARGET])
-            msg += cls._get_md5_string(
-                "-",
-                dct.get(diff.DIFF_OLD_FILE),
-                dct.get(diff.DIFF_OLD_CHECKSUM),
-            )
-            msg += cls._get_md5_string(
-                "+",
-                dct.get(diff.DIFF_NEW_FILE),
-                dct.get(diff.DIFF_NEW_CHECKSUM),
-            )
-            msg += "\n"
-            msg += cls._get_royal_changes(dct)
-        logger.info(msg)
-        return msg
+            if not self.args.checksums:
+                for _, entries in diff.items():
+                    for entry in entries:
+                        del entry["checksum"]
+
+            if self.args.show_json:
+                res = json.dumps(diff)
+            else:
+                res = self._format(diff)
+
+            logger.info(res)
 
-    def run(self):
-        try:
-            msg = self.repo.diff(
-                self.args.a_ref, target=self.args.target, b_ref=self.args.b_ref
-            )
-            self._show(msg)
         except DvcException:
-            msg = "failed to get 'diff {}'"
-            args = " ".join(
-                compact([self.args.target, self.args.a_ref, self.args.b_ref])
-            )
-            msg = msg.format(args)
-            logger.exception(msg)
+            logger.exception("failed to get diff")
             return 1
         return 0
 
 
 def add_parser(subparsers, parent_parser):
     DIFF_DESCRIPTION = (
-        "Show diff of a data file or a directory that is under DVC control.\n"
-        "Some basic statistics summary, how many files were deleted/changed."
+        "Compare two different versions of your DVC project (tracked by Git)"
+        " and shows a list of paths grouped in the following categories:"
+        " added, modified, or deleted."
     )
-    DIFF_HELP = "Show a diff of a DVC controlled data file or a directory."
     diff_parser = subparsers.add_parser(
         "diff",
         parents=[parent_parser],
         description=append_doc_link(DIFF_DESCRIPTION, "diff"),
-        help=DIFF_HELP,
+        help=DIFF_DESCRIPTION,
         formatter_class=argparse.RawDescriptionHelpFormatter,
     )
     diff_parser.add_argument(
-        "-t",
-        "--target",
+        "a_ref",
         help=(
-            "Source path to a data file or directory. Default None. "
-            "If not specified, compares all files and directories "
-            "that are under DVC control in the current working space."
+            "Git reference to the old version that you want to compare"
+            " (defaults to HEAD)"
         ),
-    )
-    diff_parser.add_argument(
-        "a_ref", help="Git reference from which diff calculates"
+        nargs="?",
+        default="HEAD",
     )
     diff_parser.add_argument(
         "b_ref",
         help=(
-            "Git reference until which diff calculates, if omitted "
-            "diff shows the difference between current HEAD and a_ref"
+            "Git reference to the new version that you want to compare."
+            " (defaults to the working tree)"
         ),
         nargs="?",
     )
+    diff_parser.add_argument(
+        "--show-json",
+        help="Format the output into a JSON",
+        action="store_true",
+        default=False,
+    )
+    diff_parser.add_argument(
+        "--checksums",
+        help="Display checksums for each entry",
+        action="store_true",
+        default=False,
+    )
     diff_parser.set_defaults(func=CmdDiff)