pdf-auto-outline 0.1.0__py3-none-any.whl

pdf_auto_outline/__init__.py

@@ -0,0 +1 @@
+__version__ = '0.1.0'

pdf_auto_outline/__main__.py

@@ -0,0 +1,4 @@
+from .main import main
+
+if __name__ == '__main__':
+    main()

pdf_auto_outline/main.py

@@ -0,0 +1,283 @@
+import pymupdf.layout
+from pymupdf import Point
+from time import perf_counter
+from multiprocessing import Pool
+import os
+import subprocess
+import argparse
+
+SIOYEK = None
+
+def log(message, end='\n'):
+    if SIOYEK:
+        SIOYEK.set_status_string(message)
+    else:
+        print(message, end=end)
+
+def get_md5_hash(path):
+    import hashlib
+    m = hashlib.md5()
+    with open(path, 'rb') as f:
+        m.update(f.read())
+    return m.hexdigest()
+
+def sioyek_transfer_annots(shared_db_path, from_hash, to_hash):
+    import sqlite3
+    con = sqlite3.connect(shared_db_path)
+    try:
+        with con:
+            con.execute("UPDATE highlights SET document_path = (?) WHERE document_path == (?)", (to_hash, from_hash))
+        log(f'moved highlights from {from_hash} to {to_hash}')
+    except Exception as e:
+        log(f'failed to move highlights from {from_hash} to {to_hash}: {e}')
+
+    con.close()
+
+
+def process_pg(fpath, pg_num) -> tuple[int, list[list]]:
+    pg = pymupdf.open(fpath)[pg_num]
+    pg.get_layout()
+
+    page_toc_entries = []
+
+    def get_text(boxclass, j, pg):
+        text = pg.get_textbox(pymupdf.Rect(*j[:4])).replace('\n', ' ').strip()
+        if boxclass == 'caption':
+            a = text.find('.', 15)
+            return text[:a] or text
+        return text
+
+    page_toc_entries = [
+            [1, get_text(j[4], j, pg), pg_num+1, j[1]]
+            for j in pg.layout_information if j[4] in ('section-header', 'caption')
+    ]
+
+    return pg_num, page_toc_entries
+
+def process_pg_wrapper(args):
+    return process_pg(args[0], args[1])
+
+def generate_toc_nnet(pdfpath, worker_cnt=3) -> list:
+    doc = pymupdf.open(pdfpath)
+    pg_cnt = doc.page_count
+    pg_nums = range(pg_cnt)
+    doc.close()
+
+    try:
+        if worker_cnt < 2:
+            log('Started..')
+            count = 1
+            bar = 50
+            entries = []
+            for i in pg_nums:
+                for j in process_pg(pdfpath, i)[1]:
+                    entries.append(j)
+                progress = (count * bar) // (pg_cnt)
+                log(f"[{'='*(progress)}{' '*(bar - progress)}] {count}/{pg_cnt} pages", end='\r')
+                count += 1
+
+            return entries
+
+
+        with Pool(processes=worker_cnt) as pool:
+            log("Started..")
+            count = 1
+            bar = 50
+            results = {}
+            tasks = [(pdfpath, i) for i in pg_nums]
+            result_iter = pool.imap_unordered(process_pg_wrapper, tasks)
+            for pg_num, res in result_iter:
+                results[pg_num] = res
+
+                progress = (count * bar) // (pg_cnt)
+                log(f"[{'='*(progress)}{' '*(bar - progress)}] {count}/{pg_cnt} pages", end='\r')
+                count += 1
+    except KeyboardInterrupt:
+        log('\nCancelled')
+        exit()
+
+    log('')
+
+    return [j for i in pg_nums for j in results[i]]
+
+def align_toc_lvls(toc_entries: list) -> list:
+    # TODO: fix this spaghetti
+    import re
+    def act(lvl, current, prev): # cur prev expected lvl
+        # if current == prev - 1: # current is parent
+        if current == prev[0]: # current is sibling
+            return lvl
+        elif current == 'p5':
+            return lvl + 1
+        elif e[current] < prev[1]: # current is parent
+            return e[current]
+            # return max(1, lvl - 1)
+        else: # e[current] > prev[1]: # current is child
+            e[current] = min(lvl + 1, e[current])
+            return min(lvl + 1, e[current])
+        # else: #e[current] == prev: # current is sibling
+        #     return lvl
+
+    p1 = re.compile(r'^[A-Z\d]')
+    p2 = re.compile(r'^(Contents)|(Chapter)|(Appendix)|(Index)|(Bibliograph)|(Preface)')
+    p3 = re.compile(r'^([IVXC\d])+\.[IVXC\d]\.? \w')
+    p4 = re.compile(r'^([AIVXC\d]+\.){2}[IVXC\d]\.? \w')
+    p5 = re.compile(r'^(Fig(ure)?\.?)|(Table\.? [\dIVXC]+)')
+    p6 = re.compile(r'''\d?\s?(Introduction)|((Materials and )?Methods)|(Results)|
+                    (Discussion)|(References)|(Summary)|(Conclusion)|(Acknowledgements)
+                    ''', re.IGNORECASE)
+    p7 = re.compile(r'^\d?\s?[A-Z ]{2,}') 
+
+    e = {'p1': 1, 'p2': 1, 'p3': 2, 'p4': 3, 'p5': 5, 'p6': 1, 'p7': 1, 'l': 2,}
+
+    log('aligning levels..')
+    lvl, prev, titles, removed = 1, ('p1', 1), set(), 0
+
+    for i in range(1, len(toc_entries)):
+        title = toc_entries[i-removed][1]
+        if (not p1.match(title)) or len(title) < 4 or title in titles: #skip
+            toc_entries.pop(i-removed)
+            removed += 1
+        elif p2.match(title):
+            lvl = act(lvl, 'p2', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p2', e['p2'])
+        elif p7.match(title):
+            lvl = act(lvl, 'p7', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p7', e['p7'])
+        elif p6.match(title):
+            lvl = act(lvl, 'p6', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p6', e['p6'])
+        elif p3.match(title):
+            lvl = act(lvl, 'p3', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p3', e['p3'])
+        elif p4.match(title):
+            lvl = act(lvl, 'p4', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p4', e['p4'])
+        elif p5.match(title):
+            lvl = act(lvl, 'p5', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('p5', e['p5'])
+        else:
+            titles.add(title)
+            lvl = act(lvl, 'l', prev)
+            toc_entries[i-removed][0] = lvl
+            prev = ('l', e['l'])
+    return toc_entries
+
+def generate_txtfile(toc_entries, txtfile='outline.txt') -> str:
+    import textwrap
+    txt = textwrap.dedent("""\
+    ============================================================
+                     TABLE OF CONTENTS OUTLINE
+    4spaces/lvl text  |  pg#  |  {details dictionary} OR y-coord
+    ============================================================
+
+    """)
+    if len(toc_entries[0]) > 3:
+        txt += '\n'.join(f"{' '*4 * (i[0] - 1)}{i[1]}  |  {i[2]}  |  {i[3]}" 
+                        for i in toc_entries)
+    else:
+        txt += '\n'.join(f"{' '*4 * (i[0] - 1)}{i[1]}  |  {i[2]}"
+                        for i in toc_entries)
+
+    with open(txtfile, 'w', encoding='utf-8') as f:
+        f.write(txt)
+
+    return txtfile
+
+
+def parse_txtfile(txtfile='outline.txt', tablevel=2) -> list:
+    toc_entries = []
+    with open(txtfile) as f:
+        if f.read(1) == '=':
+            lines = f.readlines()[5:]
+        else: lines = f.read()
+
+        for i in lines:
+            i = i.replace('\t', '    '*tablevel)
+            lvl = (len(i) - len(i.lstrip())) // 4 + 1
+            a = i.lstrip().split('  |  ')
+            if len(a) < 3:
+                toc_entries.append(
+                        [lvl, a[0], int(a[1])] 
+                )
+            else:
+                toc_entries.append(
+                        [lvl, a[0], int(a[1]), eval(a[2])]
+                )
+
+    return toc_entries
+
+def embed_toc(pdfpath, toc_entries, newfile=''):
+    print(len(toc_entries))
+    doc = pymupdf.open(pdfpath)
+    doc.set_toc(toc_entries, collapse=2)
+    if newfile:
+        doc.save(newfile)
+        log(f"toc written to '{newfile}'")
+    else:
+        doc.saveIncr()
+        log(f"toc saved to '{pdfpath}'")
+
+
+
+def edit_txtfile(txtfile='outline.txt'):
+    editor = os.environ.get('EDITOR', 'notepad' if os.name == 'nt' else 'vi')
+    subprocess.run([editor, txtfile])
+
+def main():
+    parser = argparse.ArgumentParser(prog='pdfao')
+    parser.add_argument("filename", help='input pdf')
+    parser.add_argument('-s', '--straight', action='store_true', help="write toc straight to pdf; skip editing")
+    parser.add_argument('-o', '--out', type=str, metavar='<path>', help='write changes to new pdf')
+    parser.add_argument('-mp', '--multiprocess', type=int, metavar='<n>', help='spread job over n processes (faster on linux)', default=1)
+    parser.add_argument('-e', '--edit', action='store_true', help='edit pdf toc')
+    parser.add_argument('-se', '--superedit', action='store_true', help='edit pdf toc (more attibutes available)')
+    parser.add_argument('-i', '--infile', type=str, metavar='<file>', help='write toc from file to pdf')
+    parser.add_argument('-t', '--tablevel', type=int, metavar='<n>', help='tab = n toc nesting levels (default 2)', default=2)
+    parser.add_argument('--sioyek', type=str, metavar='<path>', help='for users of the Sioyek pdf viewer')
+    parser.add_argument('--version', action='version', version='%(prog)s 0.1.0')
+
+    args = parser.parse_args()
+
+    if args.sioyek:
+        from sioyek.sioyek import Sioyek
+        sioyek_path = args.sioyek[0]
+        SIOYEK = Sioyek(sioyek_path)
+        # local_db = args.sioyek[1]
+        # shared_db = args.sioyek[2]
+        # pdf_path = args.sioyek[3]
+        # from_hash = get_md5_hash(args.filename)
+
+    if args.edit or args.superedit:
+        doc = pymupdf.Document(args.filename)
+        generate_txtfile(doc.get_toc(not args.superedit))
+        edit_txtfile()
+        toc_entries = parse_txtfile(tablevel=args.tablevel)
+        embed_toc(args.filename, toc_entries, args.out)
+    elif args.infile:
+        toc_entries = parse_txtfile(args.infile, args.tablevel)
+        embed_toc(args.filename, toc_entries, args.out)
+    else: # generate toc
+        start = perf_counter()
+        toc_entries = generate_toc_nnet(args.filename, args.multiprocess)
+        end = perf_counter()
+        log(f"finished in {end - start:<4.1f} s")
+        toc_entries = align_toc_lvls(toc_entries)
+        if args.straight:
+            embed_toc(args.filename, toc_entries, args.out)
+        else:
+            generate_txtfile(toc_entries)
+            edit_txtfile()
+            toc_entries = parse_txtfile(tablevel=args.tablevel)
+            embed_toc(args.filename, toc_entries, args.out)
+
+    # if args.sioyek and not args.out:
+    #     to_hash = get_md5_hash(args.filename)
+    #     sioyek_transfer_annots(shared_db, from_hash, to_hash)
+

pdf_auto_outline-0.1.0.dist-info/METADATA

@@ -0,0 +1,46 @@
+Metadata-Version: 2.3
+Name: pdf-auto-outline
+Version: 0.1.0
+Summary: Automatically generate and edit PDF table of contents / outline
+Author: Rossikos
+Author-email: Rossikos <216631970+rossikos@users.noreply.github.com>
+Requires-Dist: pymupdf-layout>=1.26.6
+Requires-Python: >=3.13
+Description-Content-Type: text/markdown
+
+# PDF Auto Outline
+
+A simple python program to automatically generate and embed a table of contents or outline in a PDF.
+
+## Usage
+
+```
+usage: pdfao [-h] [-s] [-o <path>] [-mp <n>] [-e] [-se] [-i <file>] [-t <n>] [--sioyek <path>] [--version] filename
+
+positional arguments:
+  filename              input pdf
+
+options:
+  -h, --help            show this help message and exit
+  -s, --straight        write toc straight to pdf; skip editing
+  -o, --out <path>      write changes to new pdf
+  -mp, --multiprocess <n>
+                        spread job over n processes (faster on linux)
+  -e, --edit            edit pdf toc
+  -se, --superedit      edit pdf toc (more attibutes available)
+  -i, --infile <file>   write toc from file to pdf
+  -t, --tablevel <n>    tab = n toc nesting levels (default 2)
+  --sioyek <path>       for users of the Sioyek pdf viewer
+  --version             show program's version number and exit
+```
+
+## For Sioyek Users
+
+Example commands; add to prefs_user.config.
+
+```
+new_command _gen_toc python3 path/to/pdfao.py "%{file_path}" --sioyek path/to/sioyek -mp 4
+new_command _edit_toc python3 path/to/pdfao.py "%{file_path}" --sioyek path/to/sioyek -e
+```
+
+

pdf_auto_outline-0.1.0.dist-info/RECORD

@@ -0,0 +1,7 @@
+pdf_auto_outline/__init__.py,sha256=IMjkMO3twhQzluVTo8Z6rE7Eg-9U79_LGKMcsWLKBkY,22
+pdf_auto_outline/__main__.py,sha256=7tzuGbeA5JiJWE_g9pzlcTXSsKlR-iEXNEbdYd4jZMs,62
+pdf_auto_outline/main.py,sha256=KkLEIGCndRql55jjvAb8Y-onmkPoQwdExHcuNG3MPYw,9977
+pdf_auto_outline-0.1.0.dist-info/WHEEL,sha256=5w2T7AS2mz1-rW9CNagNYWRCaB0iQqBMYLwKdlgiR4Q,78
+pdf_auto_outline-0.1.0.dist-info/entry_points.txt,sha256=HBvhmxJs8hHqbbpJmVTbBH3xy19Hk655O_ySwFC_53w,100
+pdf_auto_outline-0.1.0.dist-info/METADATA,sha256=ZeQBy-6lWQbqhTmt2375o5c-CQymnOffWyFeL7vi3bY,1504
+pdf_auto_outline-0.1.0.dist-info/RECORD,,

pdf_auto_outline-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: uv 0.9.7
+Root-Is-Purelib: true
+Tag: py3-none-any

pdf_auto_outline-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,4 @@
+[console_scripts]
+pdf-auto-outline = pdf_auto_outline.main:main
+pdfao = pdf_auto_outline.main:main
+