rename postgrest directory sql

9 files changed, 768 insertions, 0 deletions

diff --git a/sql/README.md b/sql/README.md
new file mode 100644
index 0000000..b171614
--- /dev/null
+++ b/sql/README.md
@@ -0,0 +1,126 @@
+
+TL;DR: replace hbase with postgresql tables with REST API (http://postgrest.org)
+
+No primary storage of anything in this table. Everything should be rapidly
+re-creatable from dumps, kafka topics (compressed), CDX, petabox metadata, etc.
+This is a secondary view on all of that.
+
+## Schema
+
+    schema/database name is 'sandcrawler'
+
+    cdx: include revisits or not?
+        id: int64, PK
+        sha1hex: string, not null, index
+        cdx_sha1hex: string
+        url: string, not null
+        datetime: ISO 8601:1988 (string?), not null
+        mimetype: string
+        warc_path: string (item and filename)
+        warc_offset: i64
+        created: datetime, index (?)
+       ?crawl: string
+       ?domain: string
+
+    file_meta
+        sha1hex, string, PK
+        md5hex: string
+        sha256hex: string
+        size_bytes: i64
+        mime: string (verifying file status; optional for now?)
+
+    fatcat_file
+        sha1hex: string, PK
+        file_ident: string, index?
+        release_ident: ?
+
+    petabox
+        id: int64, PK
+        sha1hex: string, notnull, index
+        item: string, notnull
+        path: string, notnull (TODO: URL encoded? separate sub-archive path?)
+
+    grobid
+        sha1hex: string, PK
+        updated: datetime
+        grobid_version (string)
+        status_code: i32
+        status: string (JSONB?), only if status != 200
+        metadata: JSONB, title, first author, year (not for now?)
+        glutton_fatcat_release: string, index
+
+    shadow
+        sha1hex: string, PK
+        shadow_corpus: string, PK
+        shadow_id: string
+        doi: string
+        pmid: string
+        isbn13: string
+
+Alternatively, to be more like existing system could have "one big table" or
+multiple tables all with same key (sha1b32) and UNIQ. As is, every sha1 pk
+column is 40 bytes of both index and data, or 8+ GByte (combined) for each
+table with 100 million rows. using raw bytes could help, but makes all
+code/queries much trickier.
+
+Should we have "created" or "updated" timestamps on all these columns to enable
+kafka tailing?
+
+TODO:
+- how to indicate CDX sha1 vs. true sha1 mis-match? pretty rare. recrawl and delete row from `gwb_cdx`?
+- only most recent GROBID? or keep multiple versions? here and minio
+
+## Existing Stuff Sizes (estimates)
+
+     78.5G  /user/bnewbold/journal_crawl_cdx
+     19.7G  /user/bnewbold/sandcrawler/output-prod/2018-12-14-1737.00-dumpfilemeta
+      2.7G  file_hashes.tsv
+    228.5G  /user/bnewbold/sandcrawler/output-prod/2018-09-23-0405.30-dumpgrobidmetainsertable
+
+## Use Cases
+
+Core goal here is to mostly kill hbase/hadoop. What jobs are actually used there?
+
+- backfill: load in-scope (fulltext) crawl results from CDX
+    => bulk (many line) inserts
+- rowcount: "how many unique PDFs crawled?"
+    => trivial SQL query
+- status code count: "how much GROBID progress?"
+    => trivial SQL query
+- dumpungrobided: "what files still need to be processed"
+    => SQL join with a "first" on CDX side
+- dumpgrobidxml: "merge CDX/file info with extracted XML, for those that were successful"
+    => SQL dump or rowscan, then minio fetches
+
+This table is generally "single file raw fulltext metadata".
+
+"Enrichment" jobs:
+
+- GROBID
+- glutton (if not GROBID)
+- extra file metadata
+- match newly enriched files to fatcat
+
+What else?
+
+- track additional raw file metadata
+- dump all basic GROBID metadata (title, authors, year) to attempt merge/match
+
+Questions we might want to answer
+
+- total size of PDF corpus (terabytes)
+- unqiue files hit per domain
+
+## Prototype Plan
+
+- backfill all CDX crawl files (TSV transform?)
+- load full GROBID XML (both into minio and into SQL)
+- load full fatcat file dump (TSV transform)
+- load dumpfilemeta
+
+## Example Useful Lookups
+
+
+    http get :3030/cdx?url=eq.https://coleccionables.mercadolibre.com.ar/arduino-pdf_Installments_NoInterest_BestSellers_YES
+    http get :3030/file_meta?sha1hex=eq.120582c855a7cc3c70a8527c560d7f27e6027278
+
diff --git a/sql/backfill/backfill.md b/sql/backfill/backfill.md
new file mode 100644
index 0000000..f1a5f86
--- /dev/null
+++ b/sql/backfill/backfill.md
@@ -0,0 +1,135 @@
+
+SQL Backfill Notes
+-----------------------
+
+GROBID is going to be somewhat complex.
+
+TODO:
+x CDX backfill script (CDX to postgresql direct, bulk inserts, python)
+x `file_meta` bulk insert (TSV to postgresql direct, bulk upserts, python)
+x GROBID insert (python, dump TSV to minio then postgresql)
+
+## `cdx`
+
+    #cat example.cdx | rg ' 200 ' | cut -d' ' -f2,3,4,6,9,10,11
+    #cat example.cdx | rg ' 200 ' | awk '{print $6 "\t" $3 "\t" $2 "\t" $4 "\t\t" $6 "\t" $11 "\t" $9 "\t" $10}' | b32_hex.py | awk '{print $2 "\t" $3 "\t" $4 "\t" $1 "\t" $6 "\t" $7 "\t" $8}' > cdx.example.tsv
+    cat example.cdx | ./filter_transform_cdx.py > cdx.example.tsv
+
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.example.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+
+Big HDFS import:
+
+    # but actually didn't import those; don't want to need to re-import
+    hdfs dfs -get journal_crawl_cdx/*
+
+    cat citeseerx_crawl_2017.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.citeseerx_crawl_2017.tsv
+    cat gwb-pdf-20171227034923-surt-filter/* | rg ' 200 ' | ./filter_transform_cdx.py > gwb-pdf-20171227034923-surt-filter.tsv
+    cat UNPAYWALL-PDF-CRAWL-2018-07.filtered.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.UNPAYWALL-PDF-CRAWL-2018-07.filtered.tsv
+    cat MSAG-PDF-CRAWL-2017.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.MSAG-PDF-CRAWL-2017.tsv
+
+    cat CORE-UPSTREAM-CRAWL-2018-11.sorted.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.CORE-UPSTREAM-CRAWL-2018-11.sorted.tsv
+    cat DIRECT-OA-CRAWL-2019.pdfs.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.DIRECT-OA-CRAWL-2019.pdfs.tsv
+    cat DOI-LANDING-CRAWL-2018-06.200_pdf.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.DOI-LANDING-CRAWL-2018-06.200_pdf.tsv
+    cat OA-JOURNAL-TESTCRAWL-TWO-2018.pdf.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.OA-JOURNAL-TESTCRAWL-TWO-2018.pdf.tsv
+    cat SEMSCHOLAR-PDF-CRAWL-2017.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.SEMSCHOLAR-PDF-CRAWL-2017.tsv
+    cat TARGETED-PDF-CRAWL-2017.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.TARGETED-PDF-CRAWL-2017.tsv
+    cat UNPAYWALL-PDF-CRAWL-2019-04.pdfs_sorted.cdx | rg ' 200 ' | ./filter_transform_cdx.py > cdx.UNPAYWALL-PDF-CRAWL-2019-04.pdfs_sorted.tsv
+
+TODO: nasty escaping?
+
+In psql:
+
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.UNPAYWALL-PDF-CRAWL-2018-07.filtered.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # ERROR
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.MSAG-PDF-CRAWL-2017.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # ERROR
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.citeseerx_crawl_2017.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 1653840
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.CORE-UPSTREAM-CRAWL-2018-11.sorted.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 2827563
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.DIRECT-OA-CRAWL-2019.pdfs.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 10651736
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.DOI-LANDING-CRAWL-2018-06.200_pdf.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 768565
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.OA-JOURNAL-TESTCRAWL-TWO-2018.pdf.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 5310017
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.SEMSCHOLAR-PDF-CRAWL-2017.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # COPY 2219839
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.TARGETED-PDF-CRAWL-2017.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # ERROR
+    COPY cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset) FROM '/sandcrawler-db/backfill/cdx/cdx.UNPAYWALL-PDF-CRAWL-2019-04.pdfs_sorted.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # ERROR
+
+NOTE: these largely didn't work; will need to write a batch importer.
+
+Batch import process:
+
+    cat UNPAYWALL-PDF-CRAWL-2018-07.filtered.cdx MSAG-PDF-CRAWL-2017.cdx TARGETED-PDF-CRAWL-2017.cdx UNPAYWALL-PDF-CRAWL-2019-04.pdfs_sorted.cdx | ./backfill_cdx.py
+    # Done: Counter({'raw_lines': 123254127, 'total': 51365599, 'batches': 51365})
+
+## `fatcat_file`
+
+    zcat file_export.2019-07-07.json.gz | pv -l | jq -r 'select(.sha1 != null) | [.sha1, .ident, .release_ids[0]] | @tsv' | sort -S 8G | uniq -w 40 > /sandcrawler-db/backfill/fatcat_file.2019-07-07.tsv
+
+In psql:
+
+    COPY fatcat_file FROM '/sandcrawler-db/backfill/fatcat_file.2019-07-07.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # => COPY 24727350
+
+## `file_meta`
+
+    zcat /fast/download/file_export.2019-07-07.json.gz | pv -l | jq -r 'select(.md5 != null) | [.sha1, .sha256, .md5, .size, .mimetype] | @tsv' | sort -S 8G | uniq -w 40 > /sandcrawler-db/backfill/file_meta.2019-07-07.tsv
+
+In psql:
+
+    COPY file_meta FROM '/sandcrawler-db/backfill/file_meta.2019-07-07.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # -> COPY 5860092
+
+## `petabox`
+
+    zcat /fast/download/file_export.2019-07-07.json.gz | rg '//archive.org/' | pigz > /fast/download/file_export.2019-07-07.petabox.json.gz
+    zcat /fast/download/file_export.2019-07-07.petabox.json.gz | ./petabox_transform.py | sort -u -S 8G | awk '{print $3 "\t" $1 "\t" $2}' | uniq -s40 | awk '{print $2 "\t" $3 "\t" $1}' > petabox.fatcat_2019-07-07.tsv
+
+In psql:
+
+    COPY petabox FROM '/sandcrawler-db/backfill/petabox.fatcat_2019-07-07.tsv' WITH (FORMAT TEXT, DELIMITER E'\t', NULL '');
+    # -> COPY 2887834
+
+## `grobid`
+
+Quick test:
+
+    zcat /bigger/unpaywall-transfer/2019-07-17-1741.30-dumpgrobidxml/part-00000.gz | cut -f2 | head | ./backfill_grobid.py
+
+Run big batch:
+
+    ls /bigger/unpaywall-transfer/2019-07-17-1741.30-dumpgrobidxml/part*gz | parallel --progress -j8 'zcat {} | cut -f2 | ./backfill_grobid.py'
+    # [...]
+    # Done: Counter({'minio-success': 161605, 'total': 161605, 'raw_lines': 161605, 'batches': 161})
+    # [...]
+
+Was running slow with lots of iowait and 99% jdb2. This seems to be disk I/O. Going to try:
+
+    sudo mount /dev/sdc1 /sandcrawler-minio/ -o data=writeback,noatime,nobarrier
+
+    # -j8: 20+ M/s write, little jdb2
+    # -j16: 30+ M/s write, little jdb2
+    # -j12: 30+ M/s write, going with this
+
+For general use should go back to:
+
+    sudo mount /dev/sdc1 /sandcrawler-minio/ -o data=noatime
+
+    # -j4: Still pretty slow, only ~3-5 M/s disk write. jbd2 consistently at 99%, 360 K/s write
+
+## rough table sizes
+
+                              table_name                          | table_size | indexes_size | total_size 
+    --------------------------------------------------------------+------------+--------------+------------
+     "public"."cdx"                                               | 11 GB      | 8940 MB      | 20 GB
+     "public"."shadow"                                            | 8303 MB    | 7205 MB      | 15 GB
+     "public"."fatcat_file"                                       | 5206 MB    | 2094 MB      | 7300 MB
+     "public"."file_meta"                                         | 814 MB     | 382 MB       | 1196 MB
+     "public"."petabox"                                           | 403 MB     | 594 MB       | 997 MB
+    [...]
+
diff --git a/sql/backfill/backfill_cdx.py b/sql/backfill/backfill_cdx.py
new file mode 100755
index 0000000..1c452ca
--- /dev/null
+++ b/sql/backfill/backfill_cdx.py
@@ -0,0 +1,131 @@
+#!/usr/bin/env python3
+"""
+This is a "one-time" tranform helper script for CDX backfill into sandcrawler
+postgresql.
+
+Most of this file was copied from '../python/common.py'.
+"""
+
+import json, os, sys, collections
+import base64
+import psycopg2
+import psycopg2.extras
+
+NORMAL_MIME = (
+    'application/pdf',
+    'application/postscript',
+    'text/html',
+    'text/xml',
+)
+
+def normalize_mime(raw):
+    raw = raw.lower()
+    for norm in NORMAL_MIME:
+        if raw.startswith(norm):
+            return norm
+
+    # Special cases
+    if raw.startswith('application/xml'):
+        return 'text/xml'
+    if raw.startswith('application/x-pdf'):
+        return 'application/pdf'
+    return None
+
+
+def test_normalize_mime():
+    assert normalize_mime("asdf") is None
+    assert normalize_mime("application/pdf") == "application/pdf"
+    assert normalize_mime("application/pdf+journal") == "application/pdf"
+    assert normalize_mime("Application/PDF") == "application/pdf"
+    assert normalize_mime("application/p") is None
+    assert normalize_mime("application/xml+stuff") == "text/xml"
+    assert normalize_mime("application/x-pdf") == "application/pdf"
+    assert normalize_mime("application/x-html") is None
+
+def b32_hex(s):
+    s = s.strip().split()[0].lower()
+    if s.startswith("sha1:"):
+        s = s[5:]
+    if len(s) != 32:
+        return s
+    return base64.b16encode(base64.b32decode(s.upper())).lower().decode('utf-8')
+
+
+def parse_cdx_line(raw_cdx):
+
+    cdx = raw_cdx.split()
+    if len(cdx) < 11:
+        return None
+
+    surt = cdx[0]
+    dt = cdx[1]
+    url = cdx[2]
+    mime = normalize_mime(cdx[3])
+    http_status = cdx[4]
+    key = cdx[5]
+    c_size = cdx[8]
+    offset = cdx[9]
+    warc = cdx[10]
+
+    if not (key.isalnum() and c_size.isdigit() and offset.isdigit()
+            and http_status == "200" and len(key) == 32 and dt.isdigit()
+            and mime != None):
+        return None
+
+    if '-' in (surt, dt, url, mime, http_status, key, c_size, offset, warc):
+        return None
+
+    # these are the new/specific bits
+    sha1 = b32_hex(key)
+    return dict(url=url, datetime=dt, sha1hex=sha1, cdx_sha1hex=None, mimetype=mime, warc_path=warc, warc_csize=int(c_size), warc_offset=int(offset))
+
+def insert(cur, batch):
+    sql = """
+        INSERT INTO
+        cdx (url, datetime, sha1hex, mimetype, warc_path, warc_csize, warc_offset)
+        VALUES %s
+        ON CONFLICT ON CONSTRAINT cdx_pkey DO NOTHING
+        RETURNING 1;
+    """
+    batch = [(d['url'], d['datetime'], d['sha1hex'], d['mimetype'],
+              d['warc_path'], d['warc_csize'], d['warc_offset'])
+             for d in batch]
+    res = psycopg2.extras.execute_values(cur, sql, batch) # fetch=True
+    #return len(res)
+
+def stdin_to_pg():
+    # no host means it will use local domain socket by default
+    conn = psycopg2.connect(database="sandcrawler", user="postgres")
+    cur = conn.cursor()
+    counts = collections.Counter({'total': 0})
+    batch = []
+    for l in sys.stdin:
+        l = l.strip()
+        if counts['raw_lines'] > 0 and counts['raw_lines'] % 10000 == 0:
+            print("Progress: {}...".format(counts))
+        counts['raw_lines'] += 1
+        if not l:
+            continue
+        info = parse_cdx_line(l)
+        if not info:
+            continue
+        batch.append(info)
+        counts['total'] += 1
+        if len(batch) >= 1000:
+            insert(cur, batch)
+            conn.commit()
+            #counts['inserted'] += i
+            #counts['existing'] += len(batch) - i
+            batch = []
+            counts['batches'] += 1
+    if batch:
+        insert(cur, batch)
+        #counts['inserted'] += i
+        #counts['existing'] += len(batch) - i
+        batch = []
+    conn.commit()
+    cur.close()
+    print("Done: {}".format(counts))
+
+if __name__=='__main__':
+    stdin_to_pg()
diff --git a/sql/backfill/backfill_file_meta.py b/sql/backfill/backfill_file_meta.py
new file mode 100755
index 0000000..e3b40a0
--- /dev/null
+++ b/sql/backfill/backfill_file_meta.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python3
+"""
+This is a "one-time" tranform helper script for file_meta backfill into
+sandcrawler postgresql.
+
+Most of this file was copied from '../python/common.py'.
+"""
+
+import json, os, sys, collections
+import psycopg2
+import psycopg2.extras
+
+
+def insert(cur, batch):
+    sql = """
+        INSERT INTO
+        file_meta
+        VALUES %s
+        ON CONFLICT DO NOTHING;
+    """
+    res = psycopg2.extras.execute_values(cur, sql, batch)
+
+def stdin_to_pg():
+    # no host means it will use local domain socket by default
+    conn = psycopg2.connect(database="sandcrawler", user="postgres")
+    cur = conn.cursor()
+    counts = collections.Counter({'total': 0})
+    batch = []
+    for l in sys.stdin:
+        if counts['raw_lines'] > 0 and counts['raw_lines'] % 10000 == 0:
+            print("Progress: {}...".format(counts))
+        counts['raw_lines'] += 1
+        if not l.strip():
+            continue
+        info = l.split("\t")
+        if not info:
+            continue
+        assert len(info) == 5
+        info[-1] = info[-1].strip() or None
+        batch.append(info)
+        counts['total'] += 1
+        if len(batch) >= 1000:
+            insert(cur, batch)
+            conn.commit()
+            batch = []
+            counts['batches'] += 1
+    if batch:
+        insert(cur, batch)
+        batch = []
+    conn.commit()
+    cur.close()
+    print("Done: {}".format(counts))
+
+if __name__=='__main__':
+    stdin_to_pg()
diff --git a/sql/backfill/backfill_grobid.py b/sql/backfill/backfill_grobid.py
new file mode 100755
index 0000000..08fad7f
--- /dev/null
+++ b/sql/backfill/backfill_grobid.py
@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+"""
+This is a "one-time" tranform helper script for GROBID backfill into
+sandcrawler minio and postgresql.
+"""
+
+import json, os, sys, collections, io
+import base64
+import requests
+from minio import Minio
+import psycopg2
+import psycopg2.extras
+
+
+def b32_hex(s):
+    s = s.strip().split()[0].lower()
+    if s.startswith("sha1:"):
+        s = s[5:]
+    if len(s) != 32:
+        return s
+    return base64.b16encode(base64.b32decode(s.upper())).lower().decode('utf-8')
+
+def insert(cur, batch):
+    sql = """
+        INSERT INTO
+        grobid (sha1hex, grobid_version, status_code, status, fatcat_release, metadata)
+        VALUES %s
+        ON CONFLICT DO NOTHING;
+    """
+    batch = [(d['sha1hex'], d['grobid_version'], d['status_code'], d['status'], d['fatcat_release'], d['metadata'])
+             for d in batch]
+    res = psycopg2.extras.execute_values(cur, sql, batch)
+
+def stdin_to_pg():
+    mc = Minio('localhost:9000',
+        access_key=os.environ['MINIO_ACCESS_KEY'],
+        secret_key=os.environ['MINIO_SECRET_KEY'],
+        secure=False)
+    # no host means it will use local domain socket by default
+    conn = psycopg2.connect(database="sandcrawler", user="postgres")
+    cur = conn.cursor()
+    counts = collections.Counter({'total': 0})
+    batch = []
+    for l in sys.stdin:
+        if counts['raw_lines'] > 0 and counts['raw_lines'] % 10000 == 0:
+            print("Progress: {}...".format(counts))
+        counts['raw_lines'] += 1
+        l = l.strip()
+        if not l:
+            continue
+        row = json.loads(l)
+        if not row:
+            continue
+        sha1hex = b32_hex(row['pdf_hash'])
+        grobid_xml = row['tei_xml'].encode('utf-8')
+        grobid_xml_len = len(grobid_xml)
+        grobid_xml = io.BytesIO(grobid_xml)
+
+        key = "{}/{}/{}.tei.xml".format(
+            sha1hex[0:2],
+            sha1hex[2:4],
+            sha1hex)
+        mc.put_object("grobid", key, grobid_xml, grobid_xml_len,
+            content_type="application/tei+xml",
+            metadata=None)
+        counts['minio-success'] += 1
+
+        info = dict(
+            sha1hex=sha1hex,
+            grobid_version=None, # TODO
+            status_code=200,
+            status=None,
+            fatcat_release=None,
+            metadata=None,
+        )
+        batch.append(info)
+        counts['total'] += 1
+        if len(batch) >= 1000:
+            insert(cur, batch)
+            conn.commit()
+            batch = []
+            counts['batches'] += 1
+    if batch:
+        insert(cur, batch)
+        batch = []
+    conn.commit()
+    cur.close()
+    print("Done: {}".format(counts))
+
+if __name__=='__main__':
+    stdin_to_pg()
diff --git a/sql/backfill/backfill_grobid_unpaywall.py b/sql/backfill/backfill_grobid_unpaywall.py
new file mode 100755
index 0000000..58e9e3c
--- /dev/null
+++ b/sql/backfill/backfill_grobid_unpaywall.py
@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+"""
+This is a "one-time" tranform helper script for GROBID backfill into
+sandcrawler minio and postgresql.
+
+This variant of backfill_grobid.py pushes into the unpaywall bucket of
+sandcrawler-minio and doesn't push anything to sandcrawler table in general.
+"""
+
+import json, os, sys, collections, io
+import base64
+import requests
+from minio import Minio
+import psycopg2
+import psycopg2.extras
+
+
+def b32_hex(s):
+    s = s.strip().split()[0].lower()
+    if s.startswith("sha1:"):
+        s = s[5:]
+    if len(s) != 32:
+        return s
+    return base64.b16encode(base64.b32decode(s.upper())).lower().decode('utf-8')
+
+def stdin_to_minio():
+    mc = Minio('localhost:9000',
+        access_key=os.environ['MINIO_ACCESS_KEY'],
+        secret_key=os.environ['MINIO_SECRET_KEY'],
+        secure=False)
+    counts = collections.Counter({'total': 0})
+    for l in sys.stdin:
+        if counts['raw_lines'] > 0 and counts['raw_lines'] % 10000 == 0:
+            print("Progress: {}...".format(counts))
+        counts['raw_lines'] += 1
+        l = l.strip()
+        if not l:
+            continue
+        row = json.loads(l)
+        if not row:
+            continue
+        sha1hex = b32_hex(row['pdf_hash'])
+        grobid_xml = row['tei_xml'].encode('utf-8')
+        grobid_xml_len = len(grobid_xml)
+        grobid_xml = io.BytesIO(grobid_xml)
+
+        key = "grobid/{}/{}/{}.tei.xml".format(
+            sha1hex[0:2],
+            sha1hex[2:4],
+            sha1hex)
+        mc.put_object("unpaywall", key, grobid_xml, grobid_xml_len,
+            content_type="application/tei+xml",
+            metadata=None)
+        counts['minio-success'] += 1
+
+    print("Done: {}".format(counts))
+
+if __name__=='__main__':
+    stdin_to_minio()
diff --git a/sql/backfill/filter_transform_cdx.py b/sql/backfill/filter_transform_cdx.py
new file mode 100755
index 0000000..3507dfc
--- /dev/null
+++ b/sql/backfill/filter_transform_cdx.py
@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+"""
+This is a "one-time" tranform helper script for CDX backfill into sandcrawler
+postgresql.
+
+Most of this file was copied from '../python/common.py'.
+"""
+
+import json, os, sys
+import base64
+
+NORMAL_MIME = (
+    'application/pdf',
+    'application/postscript',
+    'text/html',
+    'text/xml',
+)
+
+def normalize_mime(raw):
+    raw = raw.lower()
+    for norm in NORMAL_MIME:
+        if raw.startswith(norm):
+            return norm
+
+    # Special cases
+    if raw.startswith('application/xml'):
+        return 'text/xml'
+    if raw.startswith('application/x-pdf'):
+        return 'application/pdf'
+    return None
+
+
+def test_normalize_mime():
+    assert normalize_mime("asdf") is None
+    assert normalize_mime("application/pdf") == "application/pdf"
+    assert normalize_mime("application/pdf+journal") == "application/pdf"
+    assert normalize_mime("Application/PDF") == "application/pdf"
+    assert normalize_mime("application/p") is None
+    assert normalize_mime("application/xml+stuff") == "text/xml"
+    assert normalize_mime("application/x-pdf") == "application/pdf"
+    assert normalize_mime("application/x-html") is None
+
+def b32_hex(s):
+    s = s.strip().split()[0].lower()
+    if s.startswith("sha1:"):
+        s = s[5:]
+    if len(s) != 32:
+        return s
+    return base64.b16encode(base64.b32decode(s.upper())).lower().decode('utf-8')
+
+
+def parse_cdx_line(raw_cdx):
+
+    cdx = raw_cdx.split()
+    if len(cdx) < 11:
+        return None
+
+    surt = cdx[0]
+    dt = cdx[1]
+    url = cdx[2]
+    mime = normalize_mime(cdx[3])
+    http_status = cdx[4]
+    key = cdx[5]
+    c_size = cdx[8]
+    offset = cdx[9]
+    warc = cdx[10]
+
+    if not (key.isalnum() and c_size.isdigit() and offset.isdigit()
+            and http_status == "200" and len(key) == 32 and dt.isdigit()
+            and mime != None):
+        return None
+
+    if '-' in (surt, dt, url, mime, http_status, key, c_size, offset, warc):
+        return None
+
+    # these are the new/specific bits
+    sha1 = b32_hex(key)
+    return dict(url=url, datetime=dt, sha1hex=sha1, cdx_sha1hex=None, mimetype=mime, warc_path=warc, warc_csize=int(c_size), warc_offset=int(offset))
+
+for l in sys.stdin:
+    l = l.strip()
+    if not l:
+        continue
+    info = parse_cdx_line(l)
+    if not info:
+        continue
+    print("\t".join([info['url'], info['datetime'], info['sha1hex'], info['mimetype'], info['warc_path'], str(info['warc_csize']), str(info['warc_offset'])]))
+
diff --git a/sql/backfill/petabox_transform.py b/sql/backfill/petabox_transform.py
new file mode 100755
index 0000000..b638911
--- /dev/null
+++ b/sql/backfill/petabox_transform.py
@@ -0,0 +1,24 @@
+#!/usr/bin/env python3
+
+import json, sys, os
+
+for l in sys.stdin.readlines():
+    l = l.strip()
+    if not l:
+        continue
+    r = json.loads(l)
+    if not r['sha1']:
+        continue
+    sha1hex = r['sha1']
+    for url in r['urls']:
+        u = url['url']
+        if not '//archive.org/' in u:
+            continue
+        u = u.split('/')
+        if u[2] == 'web.archive.org':
+            continue
+        #print(u)
+        assert u[2] == 'archive.org' and u[3] in ('download', 'serve')
+        item = u[4]
+        path = '/'.join(u[5:])
+        print("\t".join([item, path, sha1hex]))
diff --git a/sql/sandcrawler_schema.sql b/sql/sandcrawler_schema.sql
new file mode 100644
index 0000000..fd921ed
--- /dev/null
+++ b/sql/sandcrawler_schema.sql
@@ -0,0 +1,59 @@
+
+CREATE TABLE IF NOT EXISTS cdx (
+    url                 TEXT NOT NULL CHECK (octet_length(url) >= 1),
+    datetime            TEXT NOT NULL CHECK (octet_length(datetime) = 14),
+    sha1hex             TEXT NOT NULL CHECK (octet_length(sha1hex) = 40),
+    cdx_sha1hex         TEXT CHECK (octet_length(cdx_sha1hex) = 40),
+    mimetype            TEXT CHECK (octet_length(mimetype) >= 1),
+    warc_path           TEXT CHECK (octet_length(warc_path) >= 1),
+    warc_csize          BIGINT,
+    warc_offset         BIGINT,
+    row_created         TIMESTAMP WITH TIME ZONE DEFAULT now() NOT NULL,
+    PRIMARY KEY(url, datetime)
+);
+CREATE INDEX IF NOT EXISTS cdx_sha1hex_idx ON cdx(sha1hex);
+CREATE INDEX IF NOT EXISTS cdx_row_created_idx ON cdx(row_created);
+
+CREATE TABLE IF NOT EXISTS file_meta (
+    sha1hex             TEXT PRIMARY KEY CHECK (octet_length(sha1hex) = 40),
+    sha256hex           TEXT CHECK (octet_length(sha256hex) = 64),
+    md5hex              TEXT CHECK (octet_length(md5hex) = 32),
+    size_bytes          BIGINT,
+    mimetype            TEXT CHECK (octet_length(mimetype) >= 1)
+);
+
+CREATE TABLE IF NOT EXISTS fatcat_file (
+    sha1hex             TEXT PRIMARY KEY CHECK (octet_length(sha1hex) = 40),
+    file_ident          TEXT CHECK (octet_length(file_ident) = 26),
+    first_release_ident TEXT CHECK (octet_length(first_release_ident) = 26)
+);
+
+CREATE TABLE IF NOT EXISTS petabox (
+    item                TEXT NOT NULL CHECK (octet_length(item) >= 1),
+    path                TEXT NOT NULL CHECK (octet_length(path) >= 1),
+    sha1hex             TEXT NOT NULL CHECK (octet_length(sha1hex) = 40),
+    PRIMARY KEY(item, path)
+);
+CREATE INDEX petabox_sha1hex_idx ON petabox(sha1hex);
+
+CREATE TABLE IF NOT EXISTS grobid (
+    sha1hex             TEXT PRIMARY KEY CHECK (octet_length(sha1hex) = 40),
+    updated             TIMESTAMP WITH TIME ZONE DEFAULT now() NOT NULL,
+    grobid_version      TEXT CHECK (octet_length(grobid_version) >= 1),
+    status_code         INT NOT NULL,
+    status              TEXT CHECK (octet_length(status) >= 1),
+    fatcat_release      TEXT CHECK (octet_length(fatcat_release) = 26),
+    metadata            JSONB
+);
+-- CREATE INDEX grobid_fatcat_release_idx ON grobid(fatcat_release);
+
+CREATE TABLE IF NOT EXISTS shadow (
+    shadow_corpus       TEXT NOT NULL CHECK (octet_length(shadow_corpus) >= 1),
+    shadow_id           TEXT NOT NULL CHECK (octet_length(shadow_id) >= 1),
+    sha1hex             TEXT NOT NULL CHECK (octet_length(sha1hex) = 40),
+    doi                 TEXT CHECK (octet_length(doi) >= 1),
+    pmid                TEXT CHECK (octet_length(pmid) >= 1),
+    isbn13              TEXT CHECK (octet_length(isbn13) >= 1),
+    PRIMARY KEY(shadow_corpus, shadow_id)
+);
+CREATE INDEX shadow_sha1hex_idx ON shadow(sha1hex);