basic inscription report

src/macrogen/graph.py

@@ -315,7 +315,7 @@ def _load_from(self, load_from: Path):
         self.closure = nx.transitive_closure(self.dag)
         self._augment_details()
 
-    def node(self, spec: Union[Reference, date, str]):
+    def node(self, spec: Union[Reference, date, str], default=KeyError):
         """
         Returns a node from the graph.
         Args:
@@ -361,7 +361,10 @@ def first(iterable):
                                   or _normalize_sigil(ref.label) == norm_spec))
 
         except StopIteration:
-            raise KeyError("No node matching {!r} in the base graph.".format(spec))
+            if isinstance(default, KeyError):
+                raise KeyError("No node matching {!r} in the base graph.".format(spec))
+            else:
+                return default
 
     def nodes(self, node_str: str, check: bool = False) -> List[Node]:
         """

src/macrogen/main.py

@@ -27,15 +27,7 @@ def main(argv=sys.argv):
 
     if not options.skip_reports:
         report.write_order_xml(graphs)
-        report.report_help()
-        report.report_refs(graphs)
-        report.report_scenes(graphs)
-        report.report_missing(graphs)
-        report.report_components(graphs)
-        report.report_conflicts(graphs)
-        report.report_sources(graphs)
-        report.report_index(graphs)
-        report.report_downloads(graphs)
+        report.generate_reports(graphs)
         render_all()
 
 

src/macrogen/report.py

@@ -2,6 +2,7 @@
 import os
 from collections import defaultdict, Counter
 from datetime import date, datetime
+from functools import partial
 from itertools import chain, repeat, groupby
 from operator import itemgetter
 
@@ -16,15 +17,15 @@
 import csv
 from html import escape
 from pathlib import Path
-from typing import Iterable, List, Dict, Mapping, Tuple, Sequence, Union, Generator, Any, Optional
+from typing import Iterable, List, Dict, Mapping, Tuple, Sequence, Union, Generator, Any, Optional, Set
 
 import networkx as nx
 from pandas import DataFrame
 
 from .config import config
 from .bibliography import BiblSource
 from .graph import MacrogenesisInfo, EARLIEST, LATEST, DAY, Node, MultiEdge
-from .graphutils import pathlink, collapse_timeline, expand_edges, in_path
+from .graphutils import pathlink, collapse_timeline, expand_edges, in_path, remove_edges
 from .datings import get_datings, AbsoluteDating
 from .uris import Reference, Witness, Inscription, UnknownRef, AmbiguousRef
 from .visualize import write_dot, simplify_graph
@@ -520,15 +521,15 @@ def report_downloads(graphs: MacrogenesisInfo):
     <h4>Nodes</h4>
     <p>The <strong>nodes</strong> are either URIs or dates in ISO 8601 format. URIs of the form
        <code>faust://document/<var>scheme</var>/<var>sigil</var></code> denote a witness (document)
-       that has the identifier <var>sigil</var> in the respective identifier scheme. 
-       <code>faust://inscription/<var>scheme</var>/<var>sigil</var>/<var>id</var></code> denote an 
+       that has the identifier <var>sigil</var> in the respective identifier scheme.
+       <code>faust://inscription/<var>scheme</var>/<var>sigil</var>/<var>id</var></code> denote an
        inscription (single “writing event”) on the respective document.</p>
         <p>If some URI has a <var>scheme</var> ≠ <code>faustedition</code>, then it was not possible to map it to
        a document in the edition. You may still try the sigil with the search. Otherwise, the document can be
        displayed at <code>http://faustedition.net/document?sigil=<var>sigil</var></code>.
        </p>
     <p>Dates are always of the form YYYY-MM-DD.</p>
-    
+
     <h4>Edges</h4>
     <p>The edges have attributes that describe them further:</p>
     <table class="pure-table">
@@ -578,7 +579,7 @@ def report_downloads(graphs: MacrogenesisInfo):
           with this assertion</td>
       </tr>
     </table>
-    
+
     <h4>MacrogenesisInfo</h4>
     <p>The <a href='macrogen-info.zip'>macrogen-info.zip</a> file contains the data required to recreate the graph info
     in the <a href='https://github.com/faustedition/faust-macrogen'>faust-macrogen</a> library. To do so, run:</p>
@@ -867,7 +868,7 @@ def report_index(graphs):
                graph_options=dict(controlIconsEnabled=True, maxZoom=200))
 
 
-def report_help():
+def report_help(info: Optional[MacrogenesisInfo] = None):
     target = config.path.report_dir
 
     def demo_graph(u, v, extend=None, **edge_attr) -> nx.MultiDiGraph:
@@ -1070,8 +1071,8 @@ def report_stats(graphs: MacrogenesisInfo):
 
     # now collect some info per witness:
     for ref in refs:
-        preds = list(graphs.base.pred[ref])
-        succs = list(graphs.base.succ[ref])
+        preds = list(graphs.base.pred[ref]) if ref in graphs.base else []
+        succs = list(graphs.base.succ[ref]) if ref in graphs.base else []
 
         pred_dates = [p for p in preds if isinstance(p, date)]
         succ_dates = [s for s in succs if isinstance(s, date)]
@@ -1110,14 +1111,14 @@ def _dating_table():
     <h2>Kanten (Aussagen)</h2>
     <p>{len(edge_df)} Kanten, {(edge_df.kind != 'timeline').sum()} Datierungsaussagen:</p>
     <pre>{edge_df.kind.value_counts()}</pre>
-    <p>{edge_df.ignore.sum()} Aussagen (manuell) ignoriert, 
-    {edge_df.delete.sum()} widersprüchliche Aussagen (automatisch) ausgeschlossen 
+    <p>{edge_df.ignore.sum()} Aussagen (manuell) ignoriert,
+    {edge_df.delete.sum()} widersprüchliche Aussagen (automatisch) ausgeschlossen
     ({len(edge_df[edge_df.delete].groupby(['start', 'end']))} ohne Parallelaussagen)
     </p>
-    
+
     <h2>Absolute Datierungen</h2>
     <table class="pure-table">
-        
+
         <thead><tr><td/><th>direkt</th><th>erschlossen</th><th>angepasst</th><th>fehlend</th></tr></thead>
         <tbody>
         <tr><th>Datumsuntergrenze</th>
@@ -1138,3 +1139,47 @@ def _dating_table():
     write_html(config.path.report_dir / "stats.php", html, "Statistik")
 
     return stat, dating_stat, edge_df
+
+
+
+def report_inscriptions(info: MacrogenesisInfo):
+
+    # all documents that have inscriptions in their textual transcript
+    from .witnesses import all_documents
+    docs = all_documents()
+    docs_by_uri = {doc.uri: doc for doc in docs}
+    tt_inscriptions = {doc.uri : doc.inscriptions for doc in docs if doc.inscriptions}
+
+    # all documents that have inscriptions in the graph
+    inscriptions_from_graph = [ref for ref in info.base if isinstance(ref, Inscription)]
+    graph_inscriptions: Dict[str, Set[str]] = defaultdict(set)
+    for inscr in inscriptions_from_graph:
+        graph_inscriptions[inscr.witness.uri].add(inscr.inscription)
+
+    relevant_uris = {uri for uri in set(tt_inscriptions.keys()).union(graph_inscriptions.keys())}
+    # stripped = remove_edges(info.base, lambda _, __ ,attr: attr.get('copy'))
+
+    table = (HtmlTable()
+             .column('Dokument', lambda uri: _fmt_node(Witness.get(uri)))
+             .column('Inskriptionen Makrogenese')
+             .column('Inskriptionen Transkript'))
+
+    def uri_idx(uri):
+        wit = info.node(uri, None)
+        return wit.index if wit else 9999
+
+    for doc_uri in sorted(relevant_uris, key=uri_idx):
+        table.row((doc_uri,
+                  '<br/>'.join(wit #f'<a href="{wit.filename.stem}">{wit.inscription}</a>'
+                               for wit in graph_inscriptions[doc_uri]),
+                  '<br/>'.join(docs_by_uri[doc_uri].inscriptions) if doc_uri in docs_by_uri else ''))
+    write_html(config.path.report_dir / 'inscriptions.php',
+               table.format_table(),
+               'Inskriptionen')
+
+
+
+def generate_reports(info: MacrogenesisInfo):
+    report_functions = [fn for name, fn in globals().items() if name.startswith('report_')]
+    for report in report_functions:
+        report(info)

src/macrogen/visualize.py

@@ -233,7 +233,7 @@ def render_all(timeout=None):
     with Pool() as pool:
         global _render_queue
         dots, _render_queue = _render_queue, []
-        result = list(tqdm(pool.imap_unordered(partial(render_file_alt, timeout=timeout), dots),
+        result = list(config.progress(pool.imap_unordered(partial(render_file_alt, timeout=timeout), dots),
                            desc='Rendering', total=len(dots), unit=' SVGs'))
         not_rendered = [entry for entry in result if isinstance(entry, tuple)]
         timeout = [path for path, err in not_rendered if isinstance(err, subprocess.TimeoutExpired)]

src/macrogen/witnesses.py

@@ -1,4 +1,5 @@
 import re
+from collections import defaultdict
 from pathlib import Path
 from typing import List, Optional
 import reprlib
@@ -35,6 +36,8 @@ def faust_uri(sigil: str, idno_type: Optional[str] = None, inscription: Optional
         components.append(inscription)
     return '/'.join(components)
 
+def _ids(reference: str):
+    return [s.strip('#') for s in reference.split()]
 
 class Document:
 
@@ -47,12 +50,41 @@ def __init__(self, source: Path):
         tt_el: etree._Element = tree.find('//f:textTranscript', config.namespaces)
         if tt_el is not None:
             self.text_transcript: Path = config.path.data.joinpath(tt_el.base.replace('faust://xml/', ''), tt_el.get('uri'))
-            transcript: etree._Element = etree.parse(fspath(self.text_transcript))
+            transcript: etree._ElementTree = etree.parse(fspath(self.text_transcript))
             self.inscriptions = transcript.xpath('//tei:change[@type="segment"]/@xml:id', namespaces=config.namespaces)
         else:
             self.text_transcript = None
             self.inscriptions = []
 
+    def _verses(self, text_transcript: Optional[etree._ElementTree] = None):
+        if text_transcript is None:
+            text_transcript = etree.parse(fspath(self.text_transcript))
+
+        lines = text_transcript.xpath('//tei:l[@n]', namespaces=config.namespaces) + \
+                text_transcript.xpath('//tei:milestone[@unit="reflines"]', namespaces=config.namespaces)
+        insc_lines = defaultdict(list)
+        if self.inscriptions:
+            for line in lines:
+                prec = line.xpath('preceding::tei:milestone[@unit="stage"][1]', namespaces=config.namespaces)
+                linenos = _ids(line.get('n'))
+                if prec:
+                    for insc in _ids(prec.get('change')):
+                        insc_lines[insc].extend(linenos)
+                else:
+                    insc_lines[''].extend(linenos)
+
+                contained = line.xpath('descendant-or-self::*/@change')
+                if contained is not None:
+                    for change in contained:
+                        for insc in _ids(change):
+                            insc_lines[insc].extend(linenos)
+
+        return insc_lines
+
+
+
+
+
     @property
     def uri(self) -> str:
         return faust_uri(self.sigil)