Render results as justified tables for easier reading

dphon/cli.py

@@ -148,17 +148,16 @@ def run() -> None:
     elif args["--output-format"] == "html":
         console.record = True
         with console.capture():
-            for match in results:
-                console.print(match)
+            for doc in graph.docs:
+                for group in doc._.groups:
+                    console.print(group)
         sys.stdout.write(console.export_html())
     else:
         # use system pager by default; colorize if LESS=R
         with console.pager(styles=os.getenv("LESS", "") == "R"):
             for doc in graph.docs:
                 for group in doc._.groups:
-                    console.print(group, "\n")
-            # for match in results:
-            #     console.print(match)
+                    console.print(group)
 
 
 def setup(args: Dict) -> Language:

dphon/console.py

@@ -1,3 +1,5 @@
+# -*- coding: utf-8 -*-
+
 from typing import List, Tuple
 
 from rich.console import Console
@@ -98,7 +100,6 @@ def _mark_span(
         for i in range(len(span)):
             # gap in u: insertion in v (if not punctuation)
             if alignment[i] == self.gap_char and other_alignment[i].isalnum():
-                marked_span.append(alignment[i])
                 other_ptr += 1
                 continue
 
@@ -136,8 +137,9 @@ def _add_span_context(self, span: Span) -> Tuple[str, str]:
         Context coloration can be changed by the default theme; a dim appearance
         is used in terminals.
         """
-        context_left = (
-            f"[context]{span.doc[span.start-self.context:span.start]}[/context]"
+        context_left = span.doc[span.start - self.context : span.start]
+        context_right = span.doc[span.end : span.end + self.context]
+        return (
+            f"[context]{context_left.text.rjust(self.context, '　')}[/context]",
+            f"[context]{context_right.text.ljust(self.context, '　')}[/context]",
         )
-        context_right = f"[context]{span.doc[span.end:span.end+self.context]}[/context]"
-        return context_left, context_right

dphon/match.py

@@ -8,6 +8,7 @@
 import Levenshtein as Lev
 from rich.console import Console, ConsoleOptions, RenderResult
 from rich.padding import Padding
+from rich.table import Table
 from spacy.tokens import Span
 
 
@@ -30,24 +31,31 @@ def __rich_console__(
         self, console: Console, options: ConsoleOptions
     ) -> RenderResult:
         """Format the match for display in console."""
+        table = Table(show_header=False, box=None)
+        table.add_column("doc")
+        table.add_column("bounds")
+        table.add_column("text")
+        table.add_column("transcription")
+
         # get colorized match text and transcription
-        su, sv = console.highlighter.format_match(self)  # type: ignore
-        pu, pv = console.highlighter.transcription(self)  # type: ignore
+        su, sv = console.highlighter.format_match(self)
+        pu, pv = console.highlighter.transcribe_match(self)
+
+        # add rows for each document
+        table.add_row(self.u, f"{self.utxt.start}-{self.utxt.end-1}", su, pu)
+        table.add_row(self.v, f"{self.vtxt.start}-{self.vtxt.end-1}", sv, pv)
 
-        # add left-padding to align with match numbers, and bottom-padding
-        # so that there's a space between matches in output
-        su, sv, pu = map(lambda t: Padding(t, (0, 0, 0, 4)), (su, sv, pu))
-        pv = Padding(pv, (0, 0, 1, 4))
+        return [table]
 
-        # return everything as an iterable of renderables
-        return (
-            f"1.  [white]{self.u}[/white] ({self.utxt.start}–{self.utxt.end-1})：",
-            su,
-            pu,
-            f"2.  [white]{self.v}[/white] ({self.vtxt.start}–{self.vtxt.end-1})：",
-            sv,
-            pv,
-        )
+    @property
+    def graphic_similarity(self) -> float:
+        """Levenshtein ratio of the aligned sequences."""
+        return Lev.seqratio(self.au, self.av)
+
+    @property
+    def phonetic_similarity(self) -> float:
+        """Similarity score of the phonetic content of the sequences."""
+        return self.weight
 
     @property
     def weighted_score(self) -> float:

dphon/reuse.py

@@ -9,6 +9,7 @@
 from networkx import MultiGraph, create_empty_copy
 from rich.console import Console, ConsoleOptions, RenderResult
 from rich.progress import BarColumn, Progress, SpinnerColumn
+from rich.table import Table
 from spacy.tokens import Doc, Span
 
 from .align import Aligner
@@ -35,29 +36,32 @@ def __rich_console__(
         self, console: Console, options: ConsoleOptions
     ) -> RenderResult:
         """Format the group for display in console."""
-        render_results = []
+        table = Table(title=self.anchor_span.text, title_justify="left", show_header=False)
+        table.add_column("doc", no_wrap=True)
+        table.add_column("text")
+        table.add_column("transcription")
 
         # render the "anchor" span first (i.e., the span that all matches share)
-        render_results += [
-            f"[bold]{self.doc._.id}[/bold] ({self.start}–{self.end-1})：",
+        table.add_row(
+            f"{self.doc._.id} ({self.start}–{self.end-1})",
             console.highlighter.format_span(self.anchor_span),
             console.highlighter.transcribe_span(self.anchor_span),
-        ]
+        )
 
         # render the non-anchor spans from each match in the group
-        for i, match in enumerate(self.matches):
+        for match in self.matches:
             span = self.non_anchor_span(match)
             alignment = self.non_anchor_alignment(match)
             anchor_alignment = self.anchor_alignment(match)
-            render_results += [
-                f"{i + 1}. {span.doc._.id} ({span.start}–{span.end-1})：",
+            table.add_row(
+                f"{span.doc._.id} ({span.start}–{span.end-1})",
                 console.highlighter.format_span(
                     span, self.anchor_span, alignment, anchor_alignment
                 ),
                 console.highlighter.transcribe_span(span),
-            ]
+            )
 
-        return render_results
+        return [table]
 
     @cached_property
     def anchor_span(self) -> Span: