moving from pybtex to bibtextparser, fixed major regressions.

* only bibtex format is supported
* all tests except test_repo.py and edit test pass
* edit and update commands were not updated
* removed --format argument from export, only bibtex is supported.

.ackrc

@@ -1 +1,2 @@
 --ignore-directory=is:build
+--ignore-directory=is:pubs.egg-info

NOTES

@@ -14,13 +14,12 @@ A paper correspond to 3 files :
 
 About strings:
 --------------
-- pybtex seems to store entries as utf-8 (TODO: check)
 - so assumption is made that everything is utf-8
 - conversions are performed at print time
 
 Config values:
 --------------
-[papers]
+[pubs]
 open-cmd = open
 edit-cmd = edit
 import-copy = True

pubs/__init__.py

@@ -1 +1 @@
-__version__ = 4
+__version__ = 4

pubs/bibstruct.py

@@ -21,20 +21,25 @@ def check_citekey(citekey):
         raise ValueError("Invalid citekey: %s" % citekey)
 
 def verify_bibdata(bibdata):
-    if not hasattr(bibdata, 'entries') or len(bibdata.entries) == 0:
-        raise ValueError('no entries in the bibdata.')
-    if len(bibdata.entries) > 1:
+    if bibdata is None or len(bibdata) == 0:
+        raise ValueError('no valid bibdata')
+    if len(bibdata) > 1:
         raise ValueError('ambiguous: multiple entries in the bibdata.')
 
 def get_entry(bibdata):
     verify_bibdata(bibdata)
-    return bibdata.entries.iteritems().next()
+    for e in bibdata.items():
+        return e
 
 def extract_citekey(bibdata):
     verify_bibdata(bibdata)
     citekey, entry = get_entry(bibdata)
     return citekey
 
+def author_last(author_str):
+    """ Return the last name of the author """
+    return author_str.split(',')[0]
+
 def generate_citekey(bibdata):
     """ Generate a citekey from bib_data.
 
@@ -44,17 +49,17 @@ def generate_citekey(bibdata):
     """
     citekey, entry = get_entry(bibdata)
 
-    author_key = 'author' if 'author' in entry.persons else 'editor'
+    author_key = 'author' if 'author' in entry else 'editor'
     try:
-        first_author = entry.persons[author_key][0]
+        first_author = entry[author_key][0]
     except KeyError:
         raise ValueError(
                 'No author or editor defined: cannot generate a citekey.')
     try:
-        year = entry.fields['year']
+        year = entry['year']
     except KeyError:
         year = ''
-    citekey = u'{}{}'.format(u''.join(first_author.last()), year)
+    citekey = u'{}{}'.format(u''.join(author_last(first_author)), year)
 
     return str2citekey(citekey)
 
@@ -67,21 +72,21 @@ def extract_docfile(bibdata, remove=False):
     citekey, entry = get_entry(bibdata)
 
     try:
-        if 'file' in entry.fields:
-            field = entry.fields['file']
+        if 'file' in entry:
+            field = entry['file']
             # Check if this is mendeley specific
             for f in field.split(':'):
                 if len(f) > 0:
                     break
             if remove:
-                entry.fields.pop('file')
+                entry.pop('file')
             # This is a hck for Mendeley. Make clean
             if f[0] != '/':
                 f = '/' + f
             return f
-        if 'attachments' in entry.fields:
-            return entry.fields['attachments']
-        if 'pdf' in entry.fields:
-            return entry.fields['pdf']
+        if 'attachments' in entry:
+            return entry['attachments']
+        if 'pdf' in entry:
+            return entry['pdf']
     except (KeyError, IndexError):
         return None

pubs/commands/add_cmd.py

@@ -108,7 +108,7 @@ def command(args):
         if copy_doc is None:
             copy_doc = config().import_copy
         if copy_doc:
-            docfile = rp.databroker.copy_doc(citekey, docfile)
+            docfile = rp.databroker.add_doc(citekey, docfile)
 
     # create the paper
 

pubs/commands/attach_cmd.py

@@ -34,7 +34,7 @@ def command(args):
     try:
         document = args.document
         if copy:
-            document = rp.databroker.copy_doc(paper.citekey, document)
+            document = rp.databroker.add_doc(paper.citekey, document)
         else:
             pass # TODO warn if file does not exists
         paper.docpath = document

pubs/commands/edit_cmd.py

@@ -15,6 +15,24 @@ def parser(subparsers):
     return parser
 
 
+def edit_meta(citekey):
+    rp = repo.Repository(config())
+    coder = endecoder.EnDecoder()
+    filepath = os.path.join(rp.databroker.databroker.filebroker.metadir(), citekey+'.yaml')
+    with open(filepath) as f:
+        content = f.read()
+
+
+
+def edit_bib(citekey):
+    rp = repo.Repository(config())
+    coder = endecoder.EnDecoder()
+    filepath = os.path.join(rp.databroker.databroker.filebroker.bibdir(), citekey+'.bib')
+    with open(filepath) as f:
+        content = f.read()
+
+
+
 def command(args):
 
     ui = get_ui()
@@ -26,8 +44,7 @@ def command(args):
     if meta:
         filepath = os.path.join(rp.databroker.databroker.filebroker.metadir(), citekey+'.yaml')
     else:
-        filepath = os.path.join(rp.databroker.databroker.filebroker.bibdir(), citekey+'.bibyaml')
-
+        filepath = os.path.join(rp.databroker.databroker.filebroker.bibdir(), citekey+'.bib')
 
     with open(filepath) as f:
         content = f.read()

pubs/commands/export_cmd.py

@@ -1,8 +1,6 @@
 from __future__ import print_function
 import sys
 
-from pybtex.database import BibliographyData
-
 from .. import repo
 from ..configs import config
 from ..uis import get_ui
@@ -11,20 +9,19 @@
 def parser(subparsers):
     parser = subparsers.add_parser('export',
             help='export bibliography')
-    parser.add_argument('-f', '--bib-format', default='bibtex',
-            help='export format')
+    # parser.add_argument('-f', '--bib-format', default='bibtex',
+    #         help='export format')
     parser.add_argument('citekeys', nargs='*',
             help='one or several citekeys')
     return parser
 
 
 def command(args):
     """
-    :param bib_format       (in 'bibtex', 'yaml')
     """
+    # :param bib_format (only 'bibtex' now)
 
     ui = get_ui()
-    bib_format = args.bib_format
 
     rp = repo.Repository(config())
 
@@ -36,12 +33,12 @@ def command(args):
 
     if len(papers) == 0:
         papers = rp.all_papers()
-    bib = BibliographyData()
+    bib = {}
     for p in papers:
-        bib.add_entry(p.citekey, p.bibentry)
+        bib[p.citekey] = p.bibentry
     try:
         exporter = endecoder.EnDecoder()
-        bibdata_raw = exporter.encode_bibdata(bib, fmt=bib_format)
+        bibdata_raw = exporter.encode_bibdata(bib)
         print(bibdata_raw, end='')
     except KeyError:
         ui.error("Invalid output format: %s." % bib_format)

pubs/commands/import_cmd.py

@@ -1,8 +1,6 @@
 import os
 import datetime
 
-from pybtex.database import Entry, BibliographyData, FieldDict, Person
-
 from .. import repo
 from .. import endecoder
 from .. import bibstruct
@@ -41,8 +39,9 @@ def many_from_path(bibpath):
 
     bibpath = os.path.expanduser(bibpath)
     if os.path.isdir(bibpath):
+        print([os.path.splitext(f)[-1][1:] for f in os.listdir(bibpath)])
         all_files = [os.path.join(bibpath, f) for f in os.listdir(bibpath)
-                     if os.path.splitext(f)[-1][1:] in list(coder.decode_fmt.keys())]
+                     if os.path.splitext(f)[-1][1:] == 'bib']
     else:
         all_files = [bibpath]
 
@@ -53,10 +52,10 @@ def many_from_path(bibpath):
 
     papers = {}
     for b in biblist:
-        for k in b.entries:
+        for k in b.keys():
             try:
-                bibdata = BibliographyData()
-                bibdata.entries[k] = b.entries[k]
+                bibdata = {}
+                bibdata[k] = b[k]
 
                 papers[k] = Paper(bibdata, citekey=k)
                 papers[k].added = datetime.datetime.now()
@@ -94,7 +93,7 @@ def command(args):
                     if copy_doc is None:
                         copy_doc = config().import_copy
                     if copy_doc:
-                        docfile = rp.databroker.copy_doc(p.citekey, docfile)
+                        docfile = rp.databroker.add_doc(p.citekey, docfile)
 
                     p.docpath = docfile
                 rp.push_paper(p)

pubs/commands/list_cmd.py

@@ -1,9 +1,9 @@
 from .. import repo
 from .. import pretty
+from .. import bibstruct
 from ..configs import config
 from ..uis import get_ui
 
-
 class InvalidQuery(ValueError):
     pass
 
@@ -56,20 +56,15 @@ def _get_field_value(query_block):
     return (field, value)
 
 
-def _lower(string, lower=True):
-    if lower:
-        return string.lower()
-    else:
-        return string
-
+def _lower(s, lower=True):
+    return s.lower() if lower else s
 
 def _check_author_match(paper, query, case_sensitive=False):
     """Only checks within last names."""
-    if not 'author' in paper.bibentry.persons:
+    if not 'author' in paper.bibentry:
         return False
-    return any([query in _lower(name, lower=(not case_sensitive))
-                for p in paper.bibentry.persons['author']
-                for name in p.last()])
+    return any([query == _lower(bibstruct.author_last(p), lower=(not case_sensitive))
+                for p in paper.bibentry['author']])
 
 
 def _check_tag_match(paper, query, case_sensitive=False):
@@ -78,7 +73,7 @@ def _check_tag_match(paper, query, case_sensitive=False):
 
 
 def _check_field_match(paper, field, query, case_sensitive=False):
-    return query in _lower(paper.bibentry.fields[field],
+    return query in _lower(paper.bibentry[field],
                        lower=(not case_sensitive))
 
 
@@ -92,7 +87,7 @@ def _check_query_block(paper, query_block, case_sensitive=None):
         return _check_tag_match(paper, value, case_sensitive=case_sensitive)
     elif field == 'author':
         return _check_author_match(paper, value, case_sensitive=case_sensitive)
-    elif field in paper.bibentry.fields:
+    elif field in paper.bibentry:
         return _check_field_match(paper, field, value,
                                   case_sensitive=case_sensitive)
     else:

pubs/datacache.py

@@ -4,15 +4,15 @@
 class DataCache(object):
     """ DataCache class, provides a very similar interface as DataBroker
 
-        Has two roles : 
+        Has two roles :
         1. Provides a buffer between the commands and the hard drive.
            Until a command request a hard drive ressource, it does not touch it.
         2. Keeps a up-to-date, pickled version of the repository, to speed up things
            when they are a lot of files. Update are also done only when required.
            Changes are detected using data modification timestamps.
 
         For the moment, only (1) is implemented.
-    """ 
+    """
     def __init__(self, directory, create=False):
         self.directory = directory
         self._databroker = None
@@ -30,16 +30,16 @@ def _create(self):
 
     def pull_metadata(self, citekey):
         return self.databroker.pull_metadata(citekey)
-        
+
     def pull_bibdata(self, citekey):
         return self.databroker.pull_bibdata(citekey)
-        
+
     def push_metadata(self, citekey, metadata):
         self.databroker.push_metadata(citekey, metadata)
-    
+
     def push_bibdata(self, citekey, bibdata):
         self.databroker.push_bibdata(citekey, bibdata)
-        
+
     def push(self, citekey, metadata, bibdata):
         self.databroker.push(citekey, metadata, bibdata)
 
@@ -59,23 +59,23 @@ def listing(self, filestats=True):
     def verify(self, bibdata_raw):
         """Will return None if bibdata_raw can't be decoded"""
         return self.databroker.verify(bibdata_raw)
-    
+
     # docbroker
 
     def in_docsdir(self, docpath):
         return self.databroker.in_docsdir(docpath)
 
     def real_docpath(self, docpath):
-        return self.databroker.real_docpath(docpath)      
+        return self.databroker.real_docpath(docpath)
 
-    def copy_doc(self, citekey, source_path, overwrite=False):
+    def add_doc(self, citekey, source_path, overwrite=False):
         return self.databroker.add_doc(citekey, source_path, overwrite=overwrite)
 
     def remove_doc(self, docpath, silent=True):
         return self.databroker.remove_doc(docpath, silent=silent)
 
     def rename_doc(self, docpath, new_citekey):
-        return self.databroker.rename_doc(docpath, new_citekey)  
+        return self.databroker.rename_doc(docpath, new_citekey)
 
     # notesbroker
 
@@ -94,7 +94,7 @@ def rename_note(self, old_citekey, new_citekey):
 #     def __init__(self, cache, directory):
 #         self.cache = cache
 #         self.directory = directory
-    
+
 #     def changes(self):
 #         """ Returns the list of modified files since the last cache was saved to disk"""
 #         pass