Basic configuration setup

logging.yaml

@@ -13,7 +13,7 @@ handlers:
 root:
   handlers:
     - console
-  level: INFO
+  level: WARNING
 
 loggers:
   graph:

src/macrogen/__init__.py

@@ -1,2 +1,4 @@
-from .graph import macrogenesis_graphs, BiblSource
+from .graph import macrogenesis_graphs
+from macrogen.bibliography import BiblSource
 from .uris import Reference, Witness, AmbiguousRef, Inscription
+from .visualize import simplify_graph, write_dot, render_file, render_all

src/macrogen/bibliography.py

@@ -0,0 +1,117 @@
+import csv
+from collections import namedtuple, defaultdict
+from typing import Dict, Union, IO
+
+import requests
+from lxml import etree
+
+from .config import config
+from . import faust
+
+BibEntry = namedtuple('BibEntry', ['uri', 'citation', 'reference', 'weight'])
+
+
+def _parse_bibliography(bibxml: Union[str, IO]) -> Dict[str, BibEntry]:
+    """Parses the bibliography file at url. Returns a dictionary mapping an URI to a corresponding bibliography entry."""
+    db: Dict[str, BibEntry] = {}
+    scores = config.bibscores
+    et = etree.parse(bibxml)
+    for bib in et.xpath('//f:bib', namespaces=faust.namespaces):
+        uri = bib.get('uri')
+        citation = bib.find('f:citation', namespaces=faust.namespaces).text
+        reference = bib.find('f:reference', namespaces=faust.namespaces).text
+        db[uri] = BibEntry(uri, citation, reference, scores[uri])
+    return db
+
+
+_bib_labels = {
+    'faust://self': 'Faustedition',
+    'faust://model/inscription': 'Inskription von',
+    'faust://orphan/adoption': 'Datierungsinhalt für',
+    'faust://heuristic': 'künstliche Datierung'
+}
+
+
+class BiblSource:
+    """
+    A bibliographic source in a macrogenesis XML file.
+    """
+
+    def __init__(self, uri, detail=''):
+        """
+        Creates a bibliographic source.
+        Args:
+            uri: should be a faust://bibliography/ URI or one of the special values
+            detail: detail string like pages
+        """
+        self.uri = uri
+        if detail is None:
+            detail = ''
+        self.detail = detail
+        self.weight = config.bibliography[uri].weight if uri in config.bibliography else 1
+
+    def __eq__(self, other):
+        if isinstance(other, BiblSource):
+            return self.uri == other.uri and self.detail == other.detail
+        else:
+            return super().__eq__(other)
+
+    def __hash__(self):
+        return hash(self.uri) ^ hash(self.detail)
+
+    def __str__(self):
+        result = self.citation
+        if self.detail is not None:
+            result += '\n' + self.detail
+        return result
+
+    @property
+    def filename(self):
+        """
+        A string representation of the source (w/o detail) that is usable as part of a filename.
+        """
+        if self.uri.startswith('faust://bibliography'):
+            return self.uri.replace('faust://bibliography/', '')
+        else:
+            return self.uri.replace('faust://', '').replace('/', '-')
+
+    @property
+    def citation(self):
+        """
+        String representation of only the citation, w/o detail.
+
+        Example:
+            Bohnenkamp 19994
+        """
+        if self.uri in config.bibliography:
+            return config.bibliography[self.uri].citation
+        elif self.uri in _bib_labels:
+            return _bib_labels[self.uri]
+        else:
+            return self.filename
+
+    @property
+    def long_citation(self):
+        if self.uri in config.bibliography:
+            return config.bibliography[self.uri].reference
+        else:
+            return self.citation
+
+
+def read_scores(scorefile):
+    """
+    Parses the bibliography score file.
+
+    Returns:
+        Map uri -> score
+
+    """
+    scores = defaultdict(lambda: 1)
+    logger = config.getLogger(__name__)
+    r = csv.reader(scorefile, delimiter='\t')
+    for row in r:
+        try:
+            scores[row[0]] = int(row[1])
+        except ValueError as e:
+            logger.warning('Skipping scorefile row %s: %s', row, e)
+    return scores