add pdfminer for text extraction;

replaced all ' with "

metadata hacking

1 files changed, 23 insertions, 23 deletions

diff --git a/web.py b/web.py
index b675c2c..7f8c0e5 100644
--- a/web.py
+++ b/web.py
@@ -33,9 +33,9 @@ def do_index():
 def do_book_file(docnum):
     with index_metadata.reader() as reader:
         document = reader.stored_fields(docnum)
-        r = werkzeug.Response(open(document['path'], "r"), mimetype="application/pdf",)
-        r.headers.add('Content-Disposition', 'attachment', 
-            filename=os.path.basename(document['path']))
+        r = werkzeug.Response(open(document["path"], "r"), mimetype="application/pdf",)
+        r.headers.add("Content-Disposition", "attachment", 
+            filename=os.path.basename(document["path"]))
         return r
 
    
@@ -53,7 +53,7 @@ def pdf_to_image(filepath, page, size):
             if stdout:
                 yield stdout
     pdffile = StringIO()
-    page = pyPdf.PdfFileReader(file(filepath, 'r')).getPage(page)
+    page = pyPdf.PdfFileReader(file(filepath, "r")).getPage(page)
     out = pyPdf.PdfFileWriter()
     out.addPage(page)
     out.write(pdffile)
@@ -66,24 +66,24 @@ def pdf_to_image(filepath, page, size):
 def do_page_image(docnum,size=260):
     with index_book.reader() as reader:
         document = reader.stored_fields(docnum)
-        page = document['pagenumber'] - 1
-        return pdf_to_image(document['path'], page, size=size)
+        page = document["pagenumber"] - 1
+        return pdf_to_image(document["path"], page, size=size)
     
  
 @app.route("/book/frontpage/<int:docnum>", methods=["GET"])
 def do_book_frontpage(docnum):
     with index_metadata.reader() as reader:
         document = reader.stored_fields(docnum)
-        return pdf_to_image(document['path'], 0, 260)
+        return pdf_to_image(document["path"], 0, 260)
  
 @app.route("/page/file/<int:docnum>", methods=["GET"])
 def do_page_file(docnum):
     with index_book.reader() as reader:
         document = reader.stored_fields(docnum)
-        filepath = document['path']
-        page = document['pagenumber'] - 1
+        filepath = document["path"]
+        page = document["pagenumber"] - 1
     app.logger.debug(str_format("Extract page={page} from filepath={filepath}", page=page, filepath=filepath))
-    inputfile = pyPdf.PdfFileReader(file(filepath, 'r'))
+    inputfile = pyPdf.PdfFileReader(file(filepath, "r"))
     pdfpage = inputfile.getPage(page)
     outbuf = StringIO()
     outfile = pyPdf.PdfFileWriter()
@@ -93,7 +93,7 @@ def do_page_file(docnum):
     r = werkzeug.Response(outbuf, mimetype="application/pdf")
     client_filename = os.path.basename(filepath)[:-3]
     client_filename += str_format(u".Page-{page}.pdf", page=page)
-    r.headers.add('Content-Disposition', 'attachment', filename=client_filename)
+    r.headers.add("Content-Disposition", "attachment", filename=client_filename)
     return r
 
 class MyHtmlFormatter(highlight.HtmlFormatter):
@@ -110,10 +110,10 @@ def do_excerpt(docnum, term):
         q = q.simplify(reader)
     terms = [ text for fieldname, text in q.all_terms()
         if fieldname == "content" ]
-    excerpt = highlight.highlight(document['content'],
+    excerpt = highlight.highlight(document["content"],
                 terms,
                 analysis.StandardAnalyzer(),
-                highlight.SimpleFragmenter(),
+                highlight.ContextFragmenter(terms, surround=40),
                 MyHtmlFormatter())
     return unicode( excerpt )
 
@@ -121,7 +121,7 @@ def do_excerpt(docnum, term):
 @app.route("/search/", methods=["GET"])
 def do_search(term=None):
     if not term:
-        return flask.render_template('search.html', objects=[], term="")
+        return flask.render_template("search.html", objects=[], term="")
 
     term = term.lower()
     searcher = index_book.searcher()
@@ -138,23 +138,23 @@ def do_search(term=None):
             with index_metadata.searcher() as searcher:
                 docnum = searcher.document_number(path=filepath)
             with index_metadata.reader() as reader2:
-                title = reader2.stored_fields(docnum).get('title')
+                title = reader2.stored_fields(docnum).get("title")
             books[docnum] = {
-                'matches' : [],
-                'title':title,
-                'filename' : os.path.basename(filepath),
+                "matches" : [],
+                "title":title,
+                "filename" : os.path.basename(filepath),
             }
             for match in book[1]:
-                pagenumber = reader.stored_fields(match[0])['pagenumber']
+                pagenumber = reader.stored_fields(match[0])["pagenumber"]
                 match = (match[0], match[1], pagenumber)
-                books[docnum]['matches'].append(match)
+                books[docnum]["matches"].append(match)
 
-    return flask.render_template('search.html',
+    return flask.render_template("search.html",
                             books=books, 
                             term=term)
 
 def log_response(sender, response):
-    sender.logger.debug('Request context is about to close down.  '
-                            'Response: %s', response)
+    sender.logger.debug("Request context is about to close down.  "
+                            "Response: %s", response)
 if __name__ == "__main__":
     app.run(host="0.0.0.0", port=8000)