papers-cli 2.18.0__tar.gz → 2.18.1__tar.gz

{papers_cli-2.18.0 → papers_cli-2.18.1}/.github/workflows/tests.yml

@@ -8,12 +8,11 @@ jobs:
   build:
 
     runs-on: ubuntu-latest
-    # 3.15t is allowed to fail: PyMuPDF does not support it yet
-    continue-on-error: ${{ matrix.python-version == '3.15t' }}
     strategy:
       fail-fast: false
       matrix:
-        python-version: [ "3.9", "3.10", "3.11", "3.12", "3.13", "3.14", "3.14t", "3.15", "3.15t" ]
+        # 3.15t: add when PyMuPDF supports it (https://github.com/pymupdf/PyMuPDF/issues/5170, #113)
+        python-version: [ "3.9", "3.10", "3.11", "3.12", "3.13", "3.14", "3.14t", "3.15" ]
 
     steps:
     - uses: actions/checkout@v6

{papers_cli-2.18.0 → papers_cli-2.18.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: papers-cli
-Version: 2.18.0
+Version: 2.18.1
 Summary: utilities to keep your PDF library organized
 Author-email: Mahé Perrette <mahe.perrette@gmail.com>
 License: MIT

{papers_cli-2.18.0 → papers_cli-2.18.1}/docs/adding-entries.md

@@ -46,6 +46,25 @@ where, above, the `papers_test` directory contains a few PDF files. For each
 PDF, `papers` will attempt to extract the metadata and add the relevant file to
 the bibliography, and rename files into the files directory.
 
+The metadata of a PDF are found as follows:
+
+1. a DOI in the PDF metadata is used as is;
+2. otherwise, the DOIs printed in the text (first pages, and the last page,
+   where some publishers add a cover page) are looked up in turn, and the
+   first one whose title appears in the first pages is used: this skips the
+   DOIs of cited papers;
+3. otherwise, a crossref search on the text of the first pages, whose result
+   is also accepted only if its title appears in the text.
+
+If none matches, the PDF is not added: pass its DOI with `--doi` (see below),
+or, for a reference without DOI (older papers, books, reports...), describe it
+and attach the PDF:
+
+```
+papers add --title "Self-consistent equilibrium ocean tides" --author "Agnew, Duncan C. and Farrell, W. E." \
+    --year 1978 --journal "Geophysical Journal International" --attachment agnew1978.pdf --rename
+```
+
 ## Add an entry from its DOI
 
 If you already know the DOI of a PDF, and don't want to gamble the fulltext

{papers_cli-2.18.0 → papers_cli-2.18.1}/docs/features.md

@@ -3,7 +3,7 @@
 ## Current features
 
 - parse PDF to extract DOI
-- fetch bibtex entry from DOI (using crossref API)
+- fetch bibtex entry from DOI (using crossref API, or doi.org for DOIs registered elsewhere, e.g. DataCite)
 - fetch bibtex entry by fulltext search (using crossref API or google scholar)
 - create and maintain bibtex file
 - add entry as PDF (`papers add ...`)

{papers_cli-2.18.0 → papers_cli-2.18.1}/papers/__main__.py

@@ -19,7 +19,7 @@ from papers.encoding import parse_file, format_file, family_names, format_entrie
 from papers.config import (bcolors, Config, search_config, CONFIG_FILE, CONFIG_FILE_LOCAL,
                            DATA_DIR, CONFIG_FILE_LEGACY, CONFIG_FILE_LEGACY_XDG)
 from papers.duplicate import list_duplicates, list_uniques, edit_entries, title_id
-from papers.entries import get_entry_val, entry_content_equal
+from papers.entries import get_entry_val, entry_content_equal, entry_from_dict
 from papers.bib import (Biblio, FUZZY_RATIO, DEFAULT_SIMILARITY, entry_filecheck,
                         backupfile as backupfile_func, isvalidkey, DuplicateKeyError, clean_filesdir,
                         are_duplicates, download_url, get_biblio, scan_dir_walk)
@@ -329,7 +329,7 @@ def addcmd(parser, o, config):
             metadata = edit_entries(metadata)
             o.edit = False
 
-        entries.extend( biblio.insert_entry(metadata, **kw) )
+        entries.extend( biblio.insert_entry(entry_from_dict(metadata), rename=o.rename, copy=o.copy, **kw) )
 
 
     # several PDFs: fetch their metadata concurrently first

{papers_cli-2.18.0 → papers_cli-2.18.1}/papers/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '2.18.0'
-__version_tuple__ = version_tuple = (2, 18, 0)
+__version__ = version = '2.18.1'
+__version_tuple__ = version_tuple = (2, 18, 1)
 
-__commit_id__ = commit_id = 'g278a27c01'
+__commit_id__ = commit_id = 'g0c4d92d88'

{papers_cli-2.18.0 → papers_cli-2.18.1}/papers/extract.py

@@ -574,7 +574,7 @@ def extract_txt_metadata(txt, search_doi=True, search_fulltext=False, max_query_
                 return bibtex
             logger.warning(f'fulltext query: title not found in the text: {title[:100]}')
 
-    raise ValueError('failed to extract metadata (pass the DOI explicitly with --doi)')
+    raise ValueError('failed to extract metadata: pass the DOI with --doi, or, without DOI, add the entry with --title, --author, --year... and the PDF with --attachment')
 
 
 def _metadata_from_pdf_contents(metadata_doi, read_text, search_doi=True, search_fulltext=True, **kw):

{papers_cli-2.18.0 → papers_cli-2.18.1}/papers_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: papers-cli
-Version: 2.18.0
+Version: 2.18.1
 Summary: utilities to keep your PDF library organized
 Author-email: Mahé Perrette <mahe.perrette@gmail.com>
 License: MIT

{papers_cli-2.18.0 → papers_cli-2.18.1}/papers_cli.egg-info/scm_version.json

@@ -1,7 +1,7 @@
 {
-  "tag": "2.18.0",
+  "tag": "2.18.1",
   "distance": 0,
-  "node": "g278a27c0165766a31c0653163d2cd7032ce50385",
+  "node": "g0c4d92d882c38efd7f126f4edeadad0266f7f4b6",
   "dirty": false,
   "branch": "HEAD",
   "node_date": "2026-10-20"

{papers_cli-2.18.0 → papers_cli-2.18.1}/tests/test_add.py

@@ -625,3 +625,33 @@ class TestAddDuplicateKeepsExistingKey(BibTest):
         paperscmd(f'add {self.otherbib} --bibtex {self.mybib} -u -m a')
         bib = Biblio.load(self.mybib, '')
         self.assertEqual(sorted(e['ID'] for e in bib.entries), ['Perrette2013ESD', 'perrette_landerer2013'])
+
+
+class TestAddWithoutDoi(unittest.TestCase):
+    """an entry described on the command line (e.g. an old paper without DOI),
+    with its PDF attached"""
+
+    def setUp(self):
+        self.tmp = tempfile.TemporaryDirectory()
+        self.mybib = os.path.join(self.tmp.name, 'papers.bib')
+        self.filesdir = os.path.join(self.tmp.name, 'files')
+        self.pdf = os.path.join(self.tmp.name, 'old paper.pdf')
+        open(self.mybib, 'w').write('')
+        open(self.pdf, 'w').write('not really a pdf')
+
+    def tearDown(self):
+        self.tmp.cleanup()
+
+    def test_add_metadata_and_attachment(self):
+        paperscmd(f'add --bibtex {self.mybib} --filesdir {self.filesdir} --rename '
+                  f'--title "Self-consistent equilibrium ocean tides" --author "Agnew, Duncan C. and Farrell, W. E." '
+                  f'--year 1978 --attachment "{self.pdf}"')
+        bib = Biblio.load(self.mybib, self.filesdir)
+        self.assertEqual(len(bib.entries), 1)
+        e = bib.entries[0]
+        self.assertEqual(e['title'], 'Self-consistent equilibrium ocean tides')
+        self.assertEqual(e['year'], '1978')
+        files = bib.get_files(e)
+        self.assertEqual(len(files), 1)
+        self.assertTrue(os.path.abspath(files[0]).startswith(self.filesdir))  # --rename moved it
+        self.assertFalse(os.path.exists(self.pdf))