aselimov/pubs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Improves document handling.

Browse Source 
- configuration is now referenced in repo object,
- introduces new class PaperInRepo,
- simplifies storage of documents in metadata,
- changes a few names.
This commit is contained in:
Olivier Mangin 2013-02-21 16:07:05 +01:00
parent 49821eab51
commit 2d700073a8
6 changed files with 141 additions and 91 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
papers/commands/import_cmd.py
View File

@ -1,7 +1,3 @@
import os
import sys
import shutil
from .. import repo
from ..paper import Paper, NoDocumentFile
from .. import files
@ -27,30 +23,23 @@ def command(config, ui, bibpath, copy):
if copy is None:
copy = config.get('papers', 'import-copy')
rp = repo.Repository.from_directory()
# Get directory for document
doc_path = files.clean_path(rp.get_document_directory(config))
if not (os.path.exists(doc_path) and os.path.isdir(doc_path)):
print "Document directory %s, does not exist." % doc_path
sys.exit(1)
# Extract papers from bib
papers = Paper.many_from_path(bibpath, fatal=False)
for p in papers:
doc_file = None
try:
file_path = p.get_document_file_from_bibdata(remove=True)
if os.path.exists(file_path):
if files.check_file(file_path):
doc_file = file_path
else:
print "File does not exist for %s." % p.citekey
print("File does not exist for %s (%s)."
% (p.citekey, file_path))
except NoDocumentFile:
print "No file for %s." % p.citekey
rp.add_paper(p)
if doc_file:
if copy:
ext = os.path.splitext(doc_file)[1]
new_doc_file = os.path.join(doc_path, p.citekey + ext)
shutil.copy(doc_file, new_doc_file)
rp.import_document(p.citekey, doc_file)
else:
new_doc_file = doc_file
p.set_document(new_doc_file)
rp.add_or_update(p)
p.set_external_document(doc_file)
rp.add_or_update(p)

14
papers/commands/open_cmd.py
View File

@ -17,14 +17,10 @@ def command(config, ui, citekey):
rp = repo.Repository.from_directory()
paper = rp.paper_from_ref(citekey, fatal=True)
try:
if paper.check_file():
filepath = paper.get_file_path()
subprocess.Popen([config.get('papers', 'open-cmd'),
filepath])
print('{} opened.'.format(colored(filepath, 'filepath')))
else:
raise NoDocumentFile
filepath = paper.get_document_path()
subprocess.Popen([config.get('papers', 'open-cmd'), filepath])
print("%s opened." % colored(filepath, 'filepath'))
except NoDocumentFile:
print('{}: No document associated to this entry {}{}{}'.format(
colored('error', 'error'), colored('citekey', 'citekey')))
print("%s: No document associated to this entry %s."
% (colored('error', 'error'), colored(citekey, 'citekey')))
exit(-1)

25
papers/files.py
View File

@ -72,17 +72,15 @@ def name_from_path(fullpdfpath, verbose=False):
return name, ext
def check_file(filepath):
if not os.path.exists(filepath):
print(colored('error', 'error') +
': {} does not exists'.format(
colored(filepath, 'filepath')))
exit(-1)
if not os.path.isfile(filepath):
print(colored('error', 'error')
+ ': {} is not a file'.format(
colored(filepath, 'filepath')))
exit(-1)
def check_file(path, fail=False):
if fail:
if not os.path.exists(path):
raise(IOError, "File does not exist: %s." % path)
if not os.path.isfile(path):
raise(IOError, "%s is not a file." % path)
return True
else:
return os.path.exists(path) and os.path.isfile(path)
# yaml I/O
@ -99,7 +97,7 @@ def write_yamlfile(filepath, datamap):
def read_yamlfile(filepath):
check_file(filepath)
check_file(filepath, fail=True)
try:
with open(filepath, 'r') as f:
return yaml.load(f)
@ -131,8 +129,7 @@ def load_meta(filepath):
# specific to bibliography data
def load_externalbibfile(fullbibpath):
check_file(fullbibpath)
check_file(fullbibpath, fail=True)
filename, ext = os.path.splitext(os.path.split(fullbibpath)[1])
if ext[1:] in FORMATS.keys():
with open(fullbibpath) as f:

109
papers/paper.py
View File

@ -1,6 +1,9 @@
import os
import unicodedata
import re
from cStringIO import StringIO
import glob
from pybtex.database import Entry, BibliographyData
@ -16,9 +19,7 @@ CITEKEY_EXCLUDE_RE = re.compile('[%s]'
% re.escape(CONTROL_CHARS + CITEKEY_FORBIDDEN_CHARS))
BASE_META = {
'filename': None,
'extension': None,
'path': None,
'external-document': None,
'notes': []
}
@ -30,6 +31,31 @@ def str2citekey(s):
return key
def get_bibentry_from_file(bibfile):
"""Extract first entry (supposed to be the only one) from given file.
"""
bib_data = files.load_externalbibfile(bibfile)
first_key = bib_data.entries.keys()[0]
first_entry = bib_data.entries[first_key]
return first_key, first_entry
def get_bibentry_from_string(content):
"""Extract first entry (supposed to be the only one) from given file.
"""
bib_data = files.parse_bibdata(StringIO(content))
first_key = bib_data.entries.keys()[0]
first_entry = bib_data.entries[first_key]
return first_key, first_entry
def get_safe_metadata(metapath):
if metapath is None:
return None
else:
return files.read_yamlfile(metapath)
class NoDocumentFile(Exception):
pass
@ -69,20 +95,22 @@ class Paper(object):
# TODO add mechanism to verify keys (15/12/2012)
def has_file(self):
"""Whether there exist a document file for this entry.
"""
return self.metadata['path'] is not None
def get_file_path(self):
if self.has_file():
return self.metadata['path']
def get_external_document_path(self):
if self.metadata['external-document'] is not None:
return self.metadata['external-document']
else:
raise NoDocumentFile
def check_file(self):
path = self.get_file_path()
return os.path.exists(path) and os.path.isfile(path)
def get_document_path(self):
return self.get_external_document_path()
def set_external_document(self, docpath):
fullpdfpath = os.path.abspath(docpath)
files.check_file(fullpdfpath, fail=True)
self.metadata['external-document'] = fullpdfpath
def check_document_path(self):
return files.check_file(self.get_external_document_path())
def generate_citekey(self):
"""Generate a citekey from bib_data.
@ -105,14 +133,6 @@ class Paper(object):
citekey = u'{}{}'.format(u''.join(first_author.last()), year)
return str2citekey(citekey)
def set_document(self, docpath):
fullpdfpath = os.path.abspath(docpath)
files.check_file(fullpdfpath)
name, ext = files.name_from_path(docpath)
self.metadata['filename'] = name
self.metadata['extension'] = ext
self.metadata['path'] = fullpdfpath
def save_to_disc(self, bib_filepath, meta_filepath):
"""Creates a BibliographyData object containing a single entry and
saves it to disc.
@ -150,23 +170,11 @@ class Paper(object):
@classmethod
def load(cls, bibpath, metapath=None):
key, entry = cls.get_bibentry(bibpath)
if metapath is None:
metadata = None
else:
metadata = files.read_yamlfile(metapath)
key, entry = get_bibentry_from_file(bibpath)
metadata = get_safe_metadata(metapath)
p = Paper(bibentry=entry, metadata=metadata, citekey=key)
return p
@classmethod
def get_bibentry(cls, bibfile):
"""Extract first entry (supposed to be the only one) from given file.
"""
bib_data = files.load_externalbibfile(bibfile)
first_key = bib_data.entries.keys()[0]
first_entry = bib_data.entries[first_key]
return first_key, first_entry
@classmethod
def create_meta(cls):
return BASE_META.copy()
@ -194,3 +202,32 @@ class Paper(object):
except ValueError, e:
print "Warning, skipping paper (%s)." % e
return papers
class PaperInRepo(Paper):
def __init__(self, repo, *args, **kwargs):
Paper.__init__(self, *args, **kwargs)
self.repo = repo
def get_document_path_in_repo(self):
doc_dir = files.clean_path(self.repo.get_document_directory())
found = glob.glob(doc_dir + "/%s.*" % self.citekey)
if found:
return found[0]
else:
raise NoDocumentFile
def get_document_path(self):
try:
return self.get_document_path_in_repo()
except NoDocumentFile:
return self.get_external_document_path()
@classmethod
def load(cls, repo, bibpath, metapath=None):
key, entry = get_bibentry_from_file(bibpath)
metadata = get_safe_metadata(metapath)
p = PaperInRepo(repo, bibentry=entry, metadata=metadata,
citekey=key)
return p

14
papers/pretty.py
View File

@ -14,12 +14,14 @@ def person_repr(p):
def short_authors(bibentry):
authors = [person_repr(p) for p in bibentry.persons['author']]
if len(authors) < 3:
return ', '.join(authors)
else:
return authors[0] + (' et al.' if len(authors) > 1 else '')
try:
authors = [person_repr(p) for p in bibentry.persons['author']]
if len(authors) < 3:
return ', '.join(authors)
else:
return authors[0] + (' et al.' if len(authors) > 1 else '')
except KeyError: # When no author is defined
return ''
def bib_oneliner(bibentry):
authors = short_authors(bibentry)

47
papers/repo.py
View File

@ -1,8 +1,10 @@
from .color import colored
import os
import shutil
import files
from paper import Paper
from paper import Paper, PaperInRepo
from color import colored
import configs
ALPHABET = 'abcdefghijklmopqrstuvwxyz'
@ -14,16 +16,20 @@ DOC_DIR = 'doc'
class Repository(object):
def __init__(self):
def __init__(self, config=None):
self.papersdir = None
self.citekeys = []
if config is None:
config = configs.CONFIG
self.config = config
# loading existing papers
def paper_from_citekey(self, citekey):
"""Load a paper by its citekey from disk, if necessary."""
return Paper.load(self.path_to_paper_file(citekey, 'bib'),
metapath=self.path_to_paper_file(citekey, 'meta'))
return PaperInRepo.load(
self, self.path_to_paper_file(citekey, 'bib'),
metapath=self.path_to_paper_file(citekey, 'meta'))
def citekey_from_ref(self, ref, fatal=True):
"""Tries to get citekey from given ref.
@ -48,9 +54,11 @@ class Repository(object):
# creating new papers
# Deprecated
# TODO merge
def add_paper_from_paths(self, docpath, bibpath):
p = Paper.load(bibpath)
p.set_document(docpath)
p.set_external_document(docpath)
self.add_paper(p)
def add_paper(self, p):
@ -73,6 +81,15 @@ class Repository(object):
else:
self.save_paper(paper)
def remove(self, citekey):
self.citetekeys.remove(citekey)
paper = self.paper_from_citekey(citekey)
for f in ('bib', 'meta'):
shutil.rmtree(self.path_to_paper_file(citekey, f))
# TODO change
if paper.metadata['in-repo']:
shutil.rmtree(self.path_to_paper_file(citekey, f))
def save_paper(self, paper):
if not paper.citekey in self.citekeys:
raise(ValueError('Paper not in repository, first add it.'))
@ -117,9 +134,9 @@ class Repository(object):
else:
raise(ValueError("%s is not a valid paper file." % file_))
def get_document_directory(self, config):
if config.has_option('papers', 'document-directory'):
return config.get('papers', 'document-directory')
def get_document_directory(self):
if self.config.has_option('papers', 'document-directory'):
return self.config.get('papers', 'document-directory')
else:
return os.path.join(self.papersdir, DOC_DIR)
@ -127,6 +144,18 @@ class Repository(object):
for key in self.citekeys:
yield self.paper_from_citekey(key)
def import_document(self, citekey, doc_file):
if citekey not in self.citekeys:
raise(ValueError, "Unknown citekey: %s." % citekey)
else:
doc_path = self.get_document_directory()
if not (os.path.exists(doc_path) and os.path.isdir(doc_path)):
raise(ValueError,
"Document directory %s, does not exist." % doc_path)
ext = os.path.splitext(doc_file)[1]
new_doc_file = os.path.join(doc_path, citekey + ext)
shutil.copy(doc_file, new_doc_file)
@classmethod
def from_directory(cls, papersdir=None):
repo = cls()