Functions to handle arXiv metadata

README.md

@@ -107,7 +107,7 @@ Here are some sources of inspirations for this project :
 
 A list of ideas and TODO. Don't hesitate to give feedback on the ones you really want or to propose your owns.
 
-20. No DOI for arXiv / HAL
+20. No DOI for HAL
 30. Parameter to disable remote search
 40. Open file
 45. Doc / Man
@@ -119,3 +119,9 @@ A list of ideas and TODO. Don't hesitate to give feedback on the ones you really
 ## Issues ?
 
 * Remove the watermarks on pdf files => done, some warning in okular on generated pdf, but seems ok. Seems to be a bug in Okular.
+
+
+## Thanks
+
+* Nathan Grigg for his [arxiv2bib](https://pypi.python.org/pypi/arxiv2bib/1.0.5#downloads) python module
+* François Boulogne for his [python-bibtexparser](https://github.com/sciunto/python-bibtexparser) python module and his integration of new requested features

fetcher.py

@@ -5,6 +5,7 @@ import isbntools
 import re
 import requesocks as requests  # Requesocks is requests with SOCKS support
 import subprocess
+import arxiv2bib as arxiv_metadata
 import tools
 import params
 
@@ -178,7 +179,6 @@ def doi2Bib(doi):
 
 
 arXiv_re = re.compile(r'arXiv:\s*([\w\.\/\-]+)')
-arXiv_wo_v_re = re.compile(r'v\d+\Z')
 
 
 def findArXivId(src):
@@ -208,21 +208,19 @@ def findArXivId(src):
         # Error happened
         tools.warning(err)
         return False
-
+    else:
-    cleanID = False
+        return extractID
-    if extractID:
-        cleanID = arXiv_wo_v_re.sub('', extractID.group(1))
-    return cleanID
-
 
 
 def arXiv2Bib(arxiv):
     """Returns bibTeX string of metadata for a given arXiv id
 
     arxiv is an arxiv id
-    From : https://github.com/minad/bibsync/blob/master/lib/bibsync/actions/synchronize_metadata.rb
     """
-    arxiv = "oai:arXiv.org:"+arxiv
+    bibtex = arxiv_metadata.arxiv2bib([arxiv])
-    bibtex = ''
+    for bib in bibtex:
-
+        if isinstance(bib, arxiv_metadata.ReferenceErrorInfo):
-    return bibtex
+            continue
+        else:
+            return bib.bibtex()
+    return False

main.py

@@ -58,37 +58,52 @@ def addFile(src, filetype):
     """
     if filetype == 'article' or filetype is None:
         doi = fetcher.findDOI(src)
+    if (filetype == 'article' or filetype is None) and doi is False:
+        arxiv = fetcher.findArXivId(src)
 
-    if filetype == 'book' or (filetype is None and doi is False):
+    if filetype == 'book' or (filetype is None and doi is False and arxiv is
+                              False):
         isbn = fetcher.findISBN(src)
 
-    if doi is False and isbn is False:
+    if doi is False and isbn is False and arxiv is False:
         if filetype is None:
-            tools.warning("Could not determine the DOI or the ISBN for " +
+            tools.warning("Could not determine the DOI nor the arXiv id nor " +
-                          src+"."+"Switching to manual entry.")
+                          "the ISBN for "+src+"."+"Switching to manual entry.")
-            doi_isbn = ''
+            doi_arxiv_isbn = ''
-            while doi_isbn not in ['doi', 'isbn']:
+            while doi_arxiv_isbn not in ['doi', 'arxiv', 'isbn']:
-                doi_isbn = tools.rawInput("DOI / ISBN ? ").lower()
+                doi_arxiv_isbn = tools.rawInput("DOI / arXiv / ISBN? ").lower()
-            if doi_isbn == 'doi':
+            if doi_arxiv_isbn == 'doi':
                 doi = tools.rawInput('DOI? ')
+            elif doi_arxiv_isbn == 'arxiv':
+                arxiv = tools.rawInput('arXiv id? ')
             else:
                 isbn = tools.rawInput('ISBN? ')
         elif filetype == 'article':
-            tools.warning("Could not determine the DOI for "+src +
+            tools.warning("Could not determine the DOI nor the arXiv id for " +
-                          ", switching to manual entry.")
+                          src+", switching to manual entry.")
+            doi_arxiv = ''
+            while doi_arxiv not in ['doi', 'arxiv']:
+                doi_arxiv = tools.rawInput("DOI / arXiv? ").lower()
+            if doi_arxiv == 'doi':
                 doi = tools.rawInput('DOI? ')
+            else:
+                arxiv = tools.rawInput('arXiv id? ')
         elif filetype == 'book':
             tools.warning("Could not determine the ISBN for "+src +
                           ", switching to manual entry.")
             isbn = tools.rawInput('ISBN? ')
     elif doi is not False:
         print("DOI for "+src+" is "+doi+".")
+    elif arxiv is not False:
+        print("ArXiv id for "+src+" is "+arxiv+".")
     elif isbn is not False:
         print("ISBN for "+src+" is "+isbn+".")
 
     if doi is not False and doi != '':
         # Add extra \n for bibtexparser
         bibtex = fetcher.doi2Bib(doi).strip().replace(',', ",\n")+"\n"
+    elif arxiv is not False and arxiv != '':
+        bibtex = fetcher.arXiv2Bib(arxiv).strip().replace(',', ",\n")+"\n"
     elif isbn is not False and isbn != '':
         # Idem
         bibtex = fetcher.isbn2Bib(isbn).strip()+"\n"
@@ -163,6 +178,14 @@ def resync():
                                                      "DOI, continue anyway " +
                                                      "? [y/N]")
                             confirm = (confirm.lower() == 'y')
+                    if 'Eprint' in entry.keys():
+                        arxiv = fetcher.findArXivId(filename)
+                        if arxiv is not False and arxiv != entry['Eprint']:
+                            confirm = tools.rawInput("Found arXiv id does " +
+                                                     "not match bibtex " +
+                                                     "entry arxiv id, " +
+                                                     "continue anyway ? [y/N]")
+                            confirm = (confirm.lower() == 'y')
                     elif 'isbn' in entry.keys():
                         isbn = fetcher.findISBN(filename)
                         if isbn is not False and isbn != entry['isbn']: