[archive] implement 'archive-event' option (#5784)

With this, IDs of skipped files will no longer be written to an archive
by default. Use "archive-event": "file,skip" to restore the previous
behavior.

docs/configuration.rst

@@ -865,6 +865,25 @@ Description
     may pose a security risk.
 
 
+extractor.*.archive-event
+-------------------------
+Type
+     + ``string``
+     + ``list`` of ``strings``
+Default
+    ``"file"``
+Example
+    * ``"file,skip"``
+    * ``["file", "skip"]``
+Description
+    `Event(s) <metadata.event_>`__
+    for which IDs get written to an
+    `archive <extractor.*.archive_>`__.
+
+    Available events are:
+    ``file``, ``skip``
+
+
 extractor.*.archive-format
 --------------------------
 Type

gallery_dl/job.py

@@ -315,7 +315,7 @@ def handle_url(self, url, kwdict):
             pathfmt.build_path()
 
             if pathfmt.exists():
-                if archive:
+                if archive and self._archive_write_skip:
                     archive.add(kwdict)
                 self.handle_skip()
                 return
@@ -345,7 +345,7 @@ def handle_url(self, url, kwdict):
                 return
 
         if not pathfmt.temppath:
-            if archive:
+            if archive and self._archive_write_skip:
                 archive.add(kwdict)
             self.handle_skip()
             return
@@ -359,7 +359,7 @@ def handle_url(self, url, kwdict):
         pathfmt.finalize()
         self.out.success(pathfmt.path)
         self._skipcnt = 0
-        if archive:
+        if archive and self._archive_write_file:
             archive.add(kwdict)
         if "after" in hooks:
             for callback in hooks["after"]:
@@ -561,6 +561,16 @@ def initialize(self, kwdict=None):
             else:
                 extr.log.debug("Using download archive '%s'", archive_path)
 
+                events = cfg("archive-event")
+                if events is None:
+                    self._archive_write_file = True
+                    self._archive_write_skip = False
+                else:
+                    if isinstance(events, str):
+                        events = events.split(",")
+                    self._archive_write_file = ("file" in events)
+                    self._archive_write_skip = ("skip" in events)
+
         skip = cfg("skip", True)
         if skip:
             self._skipexc = None
@@ -676,7 +686,7 @@ def handle_url(self, url, kwdict):
             kwdict["extension"] = "jpg"
         if self.sleep:
             self.extractor.sleep(self.sleep(), "download")
-        if self.archive:
+        if self.archive and self._archive_write_skip:
             self.archive.add(kwdict)
         self.out.skip(self.pathfmt.build_filename(kwdict))