move 'cleanups' directory from notes to extra/

4 files changed, 372 insertions, 0 deletions

diff --git a/extra/cleanups/scripts/container_dupe_to_json.py b/extra/cleanups/scripts/container_dupe_to_json.py
new file mode 100755
index 00000000..2e841c69
--- /dev/null
+++ b/extra/cleanups/scripts/container_dupe_to_json.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+
+"""
+This script can be used to transform duplicate container entity rows into JSON
+objects which can be passed to the container entity merger.
+
+It is initially used to de-dupe ISSN-Ls. The script is based on
+`file_dupe_to_json.py`.
+"""
+
+import json, sys
+from typing import Optional
+
+EXTID_TYPE = "issnl"
+
+
+def print_group(extid, dupe_ids):
+    if len(dupe_ids) < 2:
+        return
+    group = dict(
+        entity_type="container",
+        primary_id=None,
+        duplicate_ids=dupe_ids,
+        evidence=dict(
+            extid=extid,
+            extid_type=EXTID_TYPE,
+        ),
+    )
+    print(json.dumps(group, sort_keys=True))
+
+def run():
+    last_extid = None
+    dupe_ids = []
+    for l in sys.stdin:
+        l = l.strip()
+        if not l:
+            continue
+        (row_extid, row_id) = l.split("\t")[0:2]
+        if EXTID_TYPE == "issnl":
+            assert len(row_extid) == 9
+        else:
+            raise Exception(f"extid type not supported yet: {EXTID_TYPE}")
+        if row_extid == last_extid:
+            dupe_ids.append(row_id)
+            continue
+        elif dupe_ids:
+            print_group(last_extid, dupe_ids)
+        last_extid = row_extid
+        dupe_ids = [row_id]
+    if last_extid and dupe_ids:
+        print_group(last_extid, dupe_ids)
+
+
+if __name__=="__main__":
+    run()
diff --git a/extra/cleanups/scripts/fetch_full_cdx_ts.py b/extra/cleanups/scripts/fetch_full_cdx_ts.py
new file mode 100644
index 00000000..ebcf0d62
--- /dev/null
+++ b/extra/cleanups/scripts/fetch_full_cdx_ts.py
@@ -0,0 +1,201 @@
+#!/usr/bin/env python3
+
+import sys
+import json
+import base64
+from typing import Optional, List
+
+import requests
+from requests.adapters import HTTPAdapter
+from requests.packages.urllib3.util.retry import Retry  # pylint: disable=import-error
+
+def requests_retry_session(
+    retries: int = 10,
+    backoff_factor: int = 3,
+    status_forcelist: List[int] = [500, 502, 504],
+    session: requests.Session = None,
+) -> requests.Session:
+    """
+    From: https://www.peterbe.com/plog/best-practice-with-retries-with-requests
+    """
+    session = session or requests.Session()
+    retry = Retry(
+        total=retries,
+        read=retries,
+        connect=retries,
+        backoff_factor=backoff_factor,
+        status_forcelist=status_forcelist,
+    )
+    adapter = HTTPAdapter(max_retries=retry)
+    session.mount("http://", adapter)
+    session.mount("https://", adapter)
+    return session
+
+def b32_hex(s: str) -> str:
+    """
+    Converts a base32-encoded SHA-1 checksum into hex-encoded
+
+    base32 checksums are used by, eg, heritrix and in wayback CDX files
+    """
+    s = s.strip().split()[0].lower()
+    if s.startswith("sha1:"):
+        s = s[5:]
+    if len(s) != 32:
+        if len(s) == 40:
+            return s
+        raise ValueError("not a base-32 encoded SHA-1 hash: {}".format(s))
+    return base64.b16encode(base64.b32decode(s.upper())).lower().decode("utf-8")
+
+
+SANDCRAWLER_POSTGREST_URL = "http://wbgrp-svc506.us.archive.org:3030"
+
+def get_db_cdx(url: str, http_session) -> List[dict]:
+    resp = http_session.get(SANDCRAWLER_POSTGREST_URL + "/cdx", params=dict(url="eq." + url))
+    resp.raise_for_status()
+    rows = resp.json()
+    return rows or []
+
+CDX_API_URL = "https://web.archive.org/cdx/search/cdx"
+
+def get_api_cdx(url: str, partial_dt: str, http_session) -> Optional[dict]:
+
+    params = {
+        "url": url,
+        "from": partial_dt,
+        "to": partial_dt,
+        "matchType": "exact",
+        "output": "json",
+        "limit": 20,
+        # can't filter status because might be warc/revisit
+        #"filter": "statuscode:200",
+    }
+    resp = http_session.get(CDX_API_URL, params=params)
+    resp.raise_for_status()
+    rows = resp.json()
+
+    if not rows:
+        return None
+    #print(rows, file=sys.stderr)
+    if len(rows) < 2:
+        return None
+
+    for raw in rows[1:]:
+        record = dict(
+            surt=raw[0],
+            datetime=raw[1],
+            url=raw[2],
+            mimetype=raw[3],
+            status_code=raw[4],
+            sha1b32=raw[5],
+            sha1hex=b32_hex(raw[5]),
+        )
+        if record['url'] != url:
+            # TODO: could allow HTTP/HTTPS fuzzy match
+            print("CDX API near match: URL", file=sys.stderr)
+            continue
+        if not record['datetime'].startswith(partial_dt):
+            print(f"CDX API near match: datetime {partial_dt} {record['datetime']}", file=sys.stderr)
+            continue
+        if record['status_code'] == "200" or (record['status_code'] == '-' and record['mimetype'] == 'warc/revisit'):
+            return record
+        else:
+            print(f"CDX API near match: status  {record['status_code']}", file=sys.stderr)
+    return None
+
+def process_file(fe, session) -> dict:
+    short_urls = []
+    self_urls = dict()
+    full_urls = dict()
+    status = "unknown"
+
+    for pair in fe['urls']:
+        u = pair['url']
+        if not '://web.archive.org/web/' in u:
+            continue
+        seg = u.split('/')
+        assert seg[2] == "web.archive.org"
+        assert seg[3] == "web"
+        if not seg[4].isdigit():
+            continue
+        original_url = "/".join(seg[5:])
+        if len(seg[4]) == 12 or len(seg[4]) == 4:
+            short_urls.append(u)
+        elif len(seg[4]) == 14:
+            self_urls[original_url] = u
+        else:
+            print(f"other bogus ts: {seg[4]}", file=sys.stderr)
+            return dict(file_entity=fe, full_urls=full_urls, status="fail-bogus-ts")
+
+    if len(short_urls) == 0:
+        return dict(file_entity=fe, full_urls=[], status="skip-no-shorts")
+
+    for short in list(set(short_urls)):
+        seg = short.split('/')
+        ts = seg[4]
+        assert len(ts) in [12,4] and ts.isdigit()
+        original_url = '/'.join(seg[5:])
+
+        if short in full_urls:
+            continue
+
+        if original_url in self_urls and ts in self_urls[original_url]:
+            full_urls[short] = self_urls[original_url]
+            status = "success-self"
+            continue
+
+        cdx_row_list = get_db_cdx(original_url, http_session=session)
+        for cdx_row in cdx_row_list:
+            if cdx_row['sha1hex'] == fe['sha1'] and cdx_row['url'] == original_url and cdx_row['datetime'].startswith(ts):
+                assert len(cdx_row['datetime']) == 14 and cdx_row['datetime'].isdigit()
+                full_urls[short] = f"https://web.archive.org/web/{cdx_row['datetime']}/{original_url}"
+                status = "success-db"
+                break
+            else:
+                #print(f"cdx DB found, but no match", file=sys.stderr)
+                pass
+        cdx_row = None
+
+        if short in full_urls:
+            continue
+
+        cdx_record = None
+        try:
+            cdx_record = get_api_cdx(original_url, partial_dt=ts, http_session=session)
+        except requests.exceptions.HTTPError as e:
+            if e.response.status_code == 403:
+                return dict(file_entity=fe, full_urls=full_urls, status="fail-cdx-403")
+            else:
+                raise
+        if cdx_record:
+            if cdx_record['sha1hex'] == fe['sha1'] and cdx_record['url'] == original_url and cdx_record['datetime'].startswith(ts):
+                assert len(cdx_record['datetime']) == 14 and cdx_record['datetime'].isdigit()
+                full_urls[short] = f"https://web.archive.org/web/{cdx_record['datetime']}/{original_url}"
+                status = "success-api"
+                break
+            else:
+                print(f"cdx API found, but no match", file=sys.stderr)
+        else:
+            print(f"no CDX API record found: {original_url}", file=sys.stderr)
+
+        if short not in full_urls:
+            return dict(file_entity=fe, full_urls=full_urls, status="fail-not-found")
+
+    return dict(
+        file_entity=fe,
+        full_urls=full_urls,
+        status=status,
+    )
+
+def main():
+    session = requests_retry_session()
+    session.headers.update({
+        "User-Agent": "Mozilla/5.0 fatcat.CdxFixupBot",
+    })
+    for line in sys.stdin:
+        if not line.strip():
+            continue
+        fe = json.loads(line)
+        print(json.dumps(process_file(fe, session=session)))
+
+if __name__=="__main__":
+    main()
diff --git a/extra/cleanups/scripts/file2ingestrequest.py b/extra/cleanups/scripts/file2ingestrequest.py
new file mode 100755
index 00000000..a005837f
--- /dev/null
+++ b/extra/cleanups/scripts/file2ingestrequest.py
@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+
+from typing import Optional
+import json, sys
+
+
+def transform(row: dict) -> Optional[dict]:
+    if row.get('mimetype') not in [None, 'application/pdf']:
+        return None
+    if row.get('state') != 'active':
+        return None
+    base_url = None
+    for url in (row.get('urls') or []):
+        url = url['url']
+        if '://web.archive.org/' not in url and '://archive.org/' not in url:
+            base_url = url
+            break
+    if not base_url:
+        return None
+    if not row.get('sha1'):
+        return None
+    return dict(
+        base_url=base_url,
+        ingest_type="pdf",
+        link_source="fatcat",
+        link_source_id=f"file_{row['ident']}",
+        ingest_request_source="file-backfill",
+        ext_ids=dict(
+            sha1=row['sha1'],
+        ),
+    )
+
+
+def run():
+    for l in sys.stdin:
+        if not l.strip():
+            continue
+        row = json.loads(l)
+        request = transform(row)
+        if request:
+            print(json.dumps(request, sort_keys=True))
+
+if __name__=="__main__":
+    run()
diff --git a/extra/cleanups/scripts/file_dupe_to_json.py b/extra/cleanups/scripts/file_dupe_to_json.py
new file mode 100755
index 00000000..2064dc1c
--- /dev/null
+++ b/extra/cleanups/scripts/file_dupe_to_json.py
@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+
+"""
+This script can be used to transform duplicate file entity hash export rows
+into JSON objects which can be passed to the file entity merger.
+
+The input is expected to be a TSV with two columns: a hash value in the first
+column, and a fatcat file entity ident (in UUID format, not "fatcat ident"
+encoded) in the second column. The rows are assumed to be sorted by hash value
+(the first column), and duplicate values (same hash, differing UUID) are
+contiguous.
+
+File hashes aren't really "external identifiers" (ext_id), but we treat them as
+such here.
+
+Script is pretty simple, should be possible to copy and reuse for release,
+container, creator entity duplicates.
+"""
+
+import json, sys
+from typing import Optional
+import base64, uuid
+
+EXTID_TYPE = "sha1"
+
+def uuid2fcid(s: str) -> str:
+    """
+    Converts a uuid.UUID object to a fatcat identifier (base32 encoded string)
+    """     
+    raw = uuid.UUID(s).bytes
+    return base64.b32encode(raw)[:26].lower().decode("utf-8")
+
+def print_group(extid, dupe_ids):
+    if len(dupe_ids) < 2:
+        return
+    group = dict(
+        entity_type="file",
+        primary_id=None,
+        duplicate_ids=dupe_ids,
+        evidence=dict(
+            extid=extid,
+            extid_type=EXTID_TYPE,
+        ),
+    )
+    print(json.dumps(group, sort_keys=True))
+
+def run():
+    last_extid = None
+    dupe_ids = []
+    for l in sys.stdin:
+        l = l.strip()
+        if not l:
+            continue
+        (row_extid, row_uuid) = l.split("\t")[0:2]
+        if EXTID_TYPE == "sha1":
+            assert len(row_extid) == 40
+        else:
+            raise Exception(f"extid type not supported yet: {EXTID_TYPE}")
+        row_id = uuid2fcid(row_uuid)
+        if row_extid == last_extid:
+            dupe_ids.append(row_id)
+            continue
+        elif dupe_ids:
+            print_group(last_extid, dupe_ids)
+        last_extid = row_extid
+        dupe_ids = [row_id]
+    if last_extid and dupe_ids:
+        print_group(last_extid, dupe_ids)
+
+
+if __name__=="__main__":
+    run()