Add ISBN ocaid extractor bot

Author: davidscotson

isbnfromiabot/README.md

@@ -0,0 +1,13 @@
+A set of scripts to add isbn_13 values to editions with IA/ocaid references containing one.
+### How To Use
+```bash
+# Find Editions with IA ISBN, but no ISBN 13
+ ./find_editions_with_isbnianot13.sh /path/to/ol_dump.txt.gz /path/to/filtered_dump.txt.gz
+# Add ISBN 13s converted from the ia ocaid source
+python isbn_ia_to_13.py --dump_path=/path/to/filtered_dump.txt.gz --dry_run=<bool> --limit=<init>
+```
+If `dry_run` is True, the script will run as normal, but no changes will be saved to OpenLibrary.
+This is for debugging purposes. By default, `dry_run` is `True`.
+`limit` is the maximum number of changes to OpenLibrary that will occur before the script quits.
+By default, `limit` is set to `1`. Setting `limit` to `0` allows unlimited edits.
+A log is automatically generated whenever `isbn_ia_to_13.py` executes.

isbnfromiabot/find_editions_with_isbn_ia_not_13.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+
+if [[ -z $1 ]]
+  then
+    echo "No dump file provided"
+    exit 1
+fi
+if [[ -z $2 ]]
+  then
+    echo "No output file provided"
+    exit 1
+fi
+
+OL_DUMP=$1
+OUTPUT=$2
+
+zgrep ^/type/edition $OL_DUMP |
+	grep -E '"ia:isbn_\d{13}"' |
+	grep -v -E '"isbn_13":' |
+	grep -v -E '"isbn_10"' |
+	pv |
+	gzip > $OUTPUT

isbnfromiabot/isbn_ia_to_13.py

@@ -0,0 +1,70 @@
+"""
+BWB isbn ref to isbn 13
+NOTE: This script ideally works on an Open Library Dump that only contains editions with an BWB isbn ref and no isbn_13
+"""
+import gzip
+import json
+import re
+
+import isbnlib
+import olclient
+
+
+class ConvertISBNiato13Job(olclient.AbstractBotJob):
+    def run(self) -> None:
+        """Looks for any IA ISBN to convert to 13"""
+        self.write_changes_declaration()
+        header = {"type": 0, "key": 1, "revision": 2, "last_modified": 3, "JSON": 4}
+        comment = "extract ISBN 13 from IA source_record"
+        with gzip.open(self.args.file, "rb") as fin:
+            for row_num, row in enumerate(fin):
+                row = row.decode().split("\t")
+                _json = json.loads(row[header["JSON"]])
+                if _json["type"]["key"] != "/type/edition":
+                    continue
+
+                if hasattr(_json, "isbn_13"):
+                    # we only update editions with no existing isbn 13s (for now at least)
+                    continue
+
+                if "source_records" in _json:
+                    source_records = _json.get("source_records", None)
+                else:
+                    continue
+                regex = "ia:isbn_[0-9]{13}"
+                isbn_13 = False
+                for source_record in source_records:
+                    if re.fullmatch(regex, source_record):
+                        isbn_13 = source_record[8:]
+                        break
+
+                if not isbn_13:
+                    continue
+
+                if not isbnlib.is_isbn13(isbn_13):
+                    continue
+
+                olid = _json["key"].split("/")[-1]
+                edition = self.ol.Edition.get(olid)
+                if edition.type["key"] != "/type/edition":
+                    continue
+
+                if hasattr(edition, "isbn_13"):
+                    # don't update editions that already have an isbn 13
+                    continue
+
+                isbns_13 = [isbn_13]
+
+                setattr(edition, "isbn_13", isbns_13)
+                self.logger.info("\t".join([olid, source_record, str(isbns_13)]))
+                self.save(lambda: edition.save(comment=comment))
+
+
+if __name__ == "__main__":
+    job = ConvertISBNiato13Job()
+
+    try:
+        job.run()
+    except Exception as e:
+        job.logger.exception(e)
+        raise e

isbnfromiabot/requirements.txt

@@ -0,0 +1,2 @@
+openlibrary-client==0.0.30
+isbnlib==3.10.14