fix: improve citation logic (#578) bump:patch

Author: taprosoft

libs/kotaemon/kotaemon/indices/qa/citation_qa.py

@@ -334,11 +334,19 @@ def prepare_citations(self, answer, docs) -> tuple[list[Document], list[Document
             highlight_text = ""
 
             ss = sorted(ss, key=lambda x: x["start"])
+            last_end = 0
             text = cur_doc.text[: ss[0]["start"]]
+
             for idx, span in enumerate(ss):
-                to_highlight = cur_doc.text[span["start"] : span["end"]]
-                if len(to_highlight) > len(highlight_text):
-                    highlight_text = to_highlight
+                # prevent overlapping between span
+                span_start = max(last_end, span["start"])
+                span_end = max(last_end, span["end"])
+
+                to_highlight = cur_doc.text[span_start:span_end]
+                last_end = span_end
+
+                # append to highlight on PDF viewer
+                highlight_text += (" " if highlight_text else "") + to_highlight
 
                 span_idx = span.get("idx", None)
                 if span_idx is not None:
@@ -350,6 +358,7 @@ def prepare_citations(self, answer, docs) -> tuple[list[Document], list[Document
                 )
                 if idx < len(ss) - 1:
                     text += cur_doc.text[span["end"] : ss[idx + 1]["start"]]
+
             text += cur_doc.text[ss[-1]["end"] :]
             # add to display list
             with_citation.append(

libs/kotaemon/kotaemon/indices/qa/citation_qa_inline.py

@@ -152,6 +152,7 @@ def answer_to_citations(self, answer) -> list[InlineEvidence]:
     def replace_citation_with_link(self, answer: str):
         # Define the regex pattern to match 【number】
         pattern = r"【\d+】"
+        alternate_pattern = r"\[\d+\]"
 
         # Regular expression to match merged citations
         multi_pattern = r"【([\d,\s]+)】"
@@ -166,19 +167,22 @@ def split_citations(match):
         answer = re.sub(multi_pattern, split_citations, answer)
 
         # Find all citations in the answer
-        matches = re.finditer(pattern, answer)
+        matches = list(re.finditer(pattern, answer))
+        if not matches:
+            matches = list(re.finditer(alternate_pattern, answer))
 
         matched_citations = set()
         for match in matches:
             citation = match.group()
             matched_citations.add(citation)
 
         for citation in matched_citations:
+            citation_id = citation[1:-1]
             answer = answer.replace(
                 citation,
                 (
                     "<a href='#' class='citation' "
-                    f"id='mark-{citation[1:-1]}'>{citation}</a>"
+                    f"id='mark-{citation_id}'>【{citation_id}】</a>"
                 ),
             )
 

libs/kotaemon/kotaemon/indices/qa/utils.py

@@ -5,17 +5,38 @@ def find_text(search_span, context, min_length=5):
     sentence_list = search_span.split("\n")
     context = context.replace("\n", " ")
 
-    matches = []
+    matches_span = []
     # don't search for small text
     if len(search_span) > min_length:
         for sentence in sentence_list:
-            match = SequenceMatcher(
-                None, sentence, context, autojunk=False
-            ).find_longest_match()
-            if match.size > max(len(sentence) * 0.35, min_length):
-                matches.append((match.b, match.b + match.size))
+            match_results = SequenceMatcher(
+                None,
+                sentence,
+                context,
+                autojunk=False,
+            ).get_matching_blocks()
+
+            matched_blocks = []
+            for _, start, length in match_results:
+                if length > max(len(sentence) * 0.2, min_length):
+                    matched_blocks.append((start, start + length))
+
+            if matched_blocks:
+                start_index = min(start for start, _ in matched_blocks)
+                end_index = max(end for _, end in matched_blocks)
+                length = end_index - start_index
+
+                if length > max(len(sentence) * 0.35, min_length):
+                    matches_span.append((start_index, end_index))
+
+    if matches_span:
+        # merge all matches into one span
+        final_span = min(start for start, _ in matches_span), max(
+            end for _, end in matches_span
+        )
+        matches_span = [final_span]
 
-    return matches
+    return matches_span
 
 
 def find_start_end_phrase(

libs/ktem/ktem/assets/css/main.css

@@ -277,7 +277,6 @@ span.icon {
 }
 
 pdfjs-viewer-element {
-  height: 100vh;
   height: 100dvh;
 }
 
@@ -290,9 +289,8 @@ pdfjs-viewer-element {
   left: 0;
   top: 0;
   width: 100%;
-  height: 100%;
-  overflow: auto;
-  background-color: rgb(0, 0, 0);
+  height: 85dvh;
+  overflow: hidden;
   background-color: rgba(0, 0, 0, 0.4);
 }
 
@@ -302,7 +300,7 @@ pdfjs-viewer-element {
 
 .modal-content {
   background-color: #fefefe;
-  height: 110%;
+  height: 100%;
   display: flex;
   flex-direction: column;
 }
@@ -323,7 +321,7 @@ pdfjs-viewer-element {
 
 .modal-body {
   flex: 1;
-  overflow: auto;
+  overflow: hidden;
 }
 
 /* Switch checkbox styles */

libs/ktem/ktem/assets/js/main.js

@@ -32,7 +32,6 @@ function run() {
   globalThis.toggleChatColumn = (() => {
     /* get flex-grow value of chat_column */
     let flex_grow = conv_column.style.flexGrow;
-    console.log("chat col", flex_grow);
     if (flex_grow == '0') {
       conv_column.style.flexGrow = '1';
       conv_column.style.minWidth = default_conv_column_min_width;
@@ -95,10 +94,24 @@ function run() {
       event.preventDefault(); // Prevent the default link behavior
       var citationId = event.target.getAttribute('id');
 
-      await sleep(100); // Sleep for 500 milliseconds
+      await sleep(100); // Sleep for 100 milliseconds
+
+      // check if modal is open
+      var modal = document.getElementById("pdf-modal");
       var citation = document.querySelector('mark[id="' + citationId + '"]');
-      if (citation) {
-          citation.scrollIntoView({ behavior: 'smooth' });
+
+      if (modal.style.display == "block") {
+          // trigger on click event of PDF Preview link
+          var detail_elem = citation;
+          // traverse up the DOM tree to find the parent element with tag detail
+          while (detail_elem.tagName.toLowerCase() != "details") {
+              detail_elem = detail_elem.parentElement;
+          }
+          detail_elem.getElementsByClassName("pdf-link").item(0).click();
+      } else {
+        if (citation) {
+            citation.scrollIntoView({ behavior: 'smooth' });
+        }
       }
   }
 }

libs/ktem/ktem/assets/js/pdf_viewer.js

@@ -43,16 +43,52 @@ function onBlockLoad () {
             modal.style.position = "fixed";
             modal.style.width = "70%";
             modal.style.left = "15%";
+            modal.style.height = "100dvh";
           } else {
             modal.style.position = old_position;
             modal.style.width = old_width;
             modal.style.left = old_left;
+            modal.style.height = "85dvh";
           }
         };
     }
 
+    globalThis.compareText = (search_phrase, page_label) => {
+      var iframe = document.querySelector("#pdf-viewer").iframe;
+      var innerDoc = (iframe.contentDocument) ? iframe.contentDocument : iframe.contentWindow.document;
+
+      var query_selector = (
+        "#viewer > div[data-page-number='" +
+        page_label +
+        "'] > div.textLayer > span"
+      );
+      var page_spans = innerDoc.querySelectorAll(query_selector);
+      for (var i = 0; i < page_spans.length; i++) {
+        var span = page_spans[i];
+        if (
+          span.textContent.length > 4 &&
+          (
+            search_phrase.includes(span.textContent) ||
+            span.textContent.includes(search_phrase)
+          )
+        ) {
+          span.innerHTML = "<span class='highlight selected'>" + span.textContent + "</span>";
+        } else {
+          // if span is already highlighted, remove it
+          if (span.querySelector(".highlight")) {
+            span.innerHTML = span.textContent;
+          }
+        }
+      }
+    }
+
+    // Sleep function using Promise and setTimeout
+    function sleep(ms) {
+        return new Promise(resolve => setTimeout(resolve, ms));
+    }
+
     // Function to open modal and display PDF
-    globalThis.openModal = (event) => {
+    globalThis.openModal = async (event) => {
       event.preventDefault();
       var target = event.currentTarget;
       var src = target.getAttribute("data-src");
@@ -66,8 +102,8 @@ function onBlockLoad () {
       if (current_src != src) {
         pdfViewer.setAttribute("src", src);
       }
-      pdfViewer.setAttribute("phrase", phrase);
-      pdfViewer.setAttribute("search", search);
+      // pdfViewer.setAttribute("phrase", phrase);
+      // pdfViewer.setAttribute("search", search);
       pdfViewer.setAttribute("page", page);
 
       var scrollableDiv = document.getElementById("chat-info-panel");
@@ -80,6 +116,10 @@ function onBlockLoad () {
         info_panel.style.display = "none";
       }
       scrollableDiv.scrollTop = 0;
+
+      /* search for text inside PDF page */
+      await sleep(500);
+      compareText(search, page);
     }
 
     globalThis.assignPdfOnclickEvent = () => {
@@ -93,7 +133,6 @@ function onBlockLoad () {
     var created_modal = document.getElementById("pdf-viewer");
     if (!created_modal) {
         createModal();
-        console.log("Created modal")
     }
 
 }