Store embeddings and log during run

parse.py

@@ -1,6 +1,9 @@
 import re
 import os
+import json
 import xml.etree.ElementTree as ET
+from languagemodels.embeddings import embed
+from time import perf_counter
 
 
 def get_refs(root):
@@ -17,10 +20,7 @@ def get_refs(root):
 def get_text_content(root, strip_ref=False):
     """Return the plain text content of a node
 
-    This will:
-
-    1. Remove all notes, footnotes, and references
-    2. Remove all newlines and combine whitespace to single spaces
+    This will remove all notes, footnotes, and references
     """
 
     # Start with the root text prior to the first node
@@ -34,17 +34,16 @@ def get_text_content(root, strip_ref=False):
             # Add the root node text between this child and the next child
             text += child.tail
 
-    if strip_ref:
-        # Remove empty parens from removed references
-        text = re.sub(r"\([\n\t; ]*\)", " ", text)
+    return text
 
-        # Combine whitespace to single space
-        # text = re.sub(r"[\n\t ]+", " ", text)
 
-    return text
+log = open("log.txt", "w")
+jsonl = open("train.jsonl", "w")
+count = 0
 
 
 def get_paras():
+    global count
     paras = []
 
     from pathlib import Path
@@ -53,7 +52,6 @@ def get_paras():
         filename = str(filename)
         if "authInfo." in filename:
             continue
-        print(filename)
         try:
             tree = ET.parse(filename)
         except:
@@ -62,31 +60,39 @@ def get_paras():
 
         root = tree.getroot()
 
-        # for i, p in enumerate(root.findall(".//p[scripRef]")):
         for i, p in enumerate(root.findall(".//p")):
+            xml = ET.tostring(p, encoding="unicode")
             text = get_text_content(p)
 
-            if False:
-                text = re.sub(r'".*?"', "...", text, flags=re.M | re.DOTALL)
-                text = re.sub(r"“.*?”", "...", text, flags=re.M | re.DOTALL)
-
-                # Remove spaces before punctuation
-                text = re.sub(r"[\t\n ]\.", ".", text)
-                text = re.sub(r"[\t\n ],", ",", text)
-
-                # Combine elipses
-                text = re.sub(r"\.\.\.+", "...", text)
-
-            refs = get_refs(p)
+            if "id" in p.attrib:
+                start = perf_counter()
+                refs = get_refs(p)
+                emb = list(float(n) for n in embed([text])[0])
+                count += 1
 
-            if len(text) > 160 and "id" in p.attrib:
-                yield {
+                row = {
                     "id": filename + ":" + p.attrib["id"],
-                    "refs": refs,
                     "text": text,
-                    "text-noref": get_text_content(p, strip_ref=True),
+                    "thml": xml,
+                    "refs": refs,
+                    "all-MiniLM-L6-v2": emb,
                 }
 
+                print(
+                    count,
+                    f"{perf_counter()-start:.2f}",
+                    len(row["text"]),
+                    len(row["thml"]),
+                    row["id"],
+                    row["all-MiniLM-L6-v2"][:2],
+                    file=log,
+                    flush=True,
+                )
+
+                print(json.dumps(row), file=jsonl, flush=True)
+
+                yield row
+
 
 from datasets import Dataset