Edit book/Book polishing: When embedding fonts or subsetting embedded fonts, handle the :first-letter and :first-line pseudo-element style rules. Fixes #1373649 [Book Editor - Font Subsetting does not recognize pseudo-element declarations](https://bugs.launchpad.net/calibre/+bug/1373649)

src/calibre/ebooks/oeb/polish/font_stats.coffee

@@ -50,6 +50,29 @@ process_font_face_rule = (rule, font_faces) ->
     fd['src'] = rule.style.getPropertyValue('src')
     font_faces.push(fd)
 
+fl_pat = /:{1,2}(first-letter|first-line)/i
+
+process_sheet_for_pseudo = (sheet, rules) ->
+    for rule in sheet.cssRules
+        if rule.type == rule.STYLE_RULE
+            st = rule.selectorText
+            m = fl_pat.exec(st)
+            if m
+                pseudo = m[1].toLowerCase()
+                ff = rule.style.getPropertyValue('font-family')
+                if ff
+                    process_style_rule(st, rule.style, rules, pseudo)
+        else if rule.type == rule.IMPORT_RULE and rule.styleSheet
+            process_sheet_for_pseudo(rule.styleSheet, rules)
+
+process_style_rule = (selector_text, style, rules, pseudo) ->
+    selector_text = selector_text.replace(fl_pat, '')
+    fd = font_dict(style)
+    for element in document.querySelectorAll(selector_text)
+        text = element.innerText
+        if text
+            rules.push([fd, text, pseudo])
+
 class FontStats
     # This class is a namespace to expose functions via the
     # window.font_stats object.
@@ -75,6 +98,12 @@ class FontStats
                 ans.push(usage)
         py_bridge.value = ans
 
+    get_pseudo_element_font_usage: () ->
+        ans = []
+        for sheet in document.styleSheets
+            process_sheet_for_pseudo(sheet, ans)
+        py_bridge.value = ans
+
     get_font_families: () ->
         ans = {}
         for node in document.getElementsByTagName('*')

src/calibre/ebooks/oeb/polish/stats.py

@@ -11,6 +11,7 @@ import json, sys, os, logging
 from urllib import unquote
 from collections import defaultdict
 
+import regex
 from cssutils import CSSParser
 from PyQt5.Qt import (pyqtProperty, QEventLoop, Qt, QSize, QTimer,
                       pyqtSlot)
@@ -101,6 +102,31 @@ def get_matching_rules(rules, font):
             return m
     return []
 
+def parse_font_families(parser, raw):
+    style = parser.parseStyle('font-family:' + raw, validate=False).getProperty('font-family')
+    for x in style.propertyValue:
+        x = x.value
+        if x:
+            yield x
+
+def get_pseudo_element_font_usage(pseudo_element_font_usage, first_letter_pat, parser):
+    ans = []
+    for font_dict, text, pseudo in pseudo_element_font_usage:
+        text = text.strip()
+        if pseudo == 'first-letter':
+            prefix = first_letter_pat.match(text)
+            if prefix is not None:
+                text = prefix + text[len(prefix):].lstrip()[:1]
+            else:
+                text = text[:1]
+        if text:
+            font = font_dict.copy()
+            font['text'] = text
+            font['font-family'] = list(parse_font_families(parser, font['font-family']))
+            ans.append(font)
+
+    return ans
+
 class Page(QWebPage):  # {{{
 
     def __init__(self, log):
@@ -164,6 +190,7 @@ class StatsCollector(object):
         self.do_embed = do_embed
         must_use_qt()
         self.parser = CSSParser(loglevel=logging.CRITICAL, log=logging.getLogger('calibre.css'))
+        self.first_letter_pat = regex.compile(r'^[\p{Ps}\p{Ps}\p{Pe}\p{Pi}\p{Pf}\p{Po}]+', regex.VERSION1 | regex.UNICODE)
 
         self.loop = QEventLoop()
         self.view = QWebView()
@@ -186,6 +213,14 @@ class StatsCollector(object):
         if self.loop.exec_() == 1:
             raise Exception('Failed to gather statistics from book, see log for details')
 
+    def log_exception(self, *args):
+        orig = self.log.filter_level
+        try:
+            self.log.filter_level = self.log.DEBUG
+            self.log.exception(*args)
+        finally:
+            self.log.filter_level = orig
+
     def render_book(self):
         try:
             if not self.render_queue:
@@ -193,7 +228,7 @@ class StatsCollector(object):
             else:
                 self.render_next()
         except:
-            self.logger.exception('Rendering failed')
+            self.log_exception('Rendering failed')
             self.loop.exit(1)
 
     def render_next(self):
@@ -210,7 +245,7 @@ class StatsCollector(object):
             self.page.load_js()
             self.collect_font_stats()
         except:
-            self.log.exception('Failed to collect font stats from: %s'%self.container.relpath(self.current_item))
+            self.log_exception('Failed to collect font stats from: %s'%self.container.relpath(self.current_item))
             self.loop.exit(1)
             return
 
@@ -283,6 +318,11 @@ class StatsCollector(object):
         font_usage = self.page.bridge_value
         if not isinstance(font_usage, list):
             raise Exception('Unknown error occurred while reading font usage')
+        self.page.evaljs('window.font_stats.get_pseudo_element_font_usage()')
+        pseudo_element_font_usage = self.page.bridge_value
+        if not isinstance(pseudo_element_font_usage, list):
+            raise Exception('Unknown error occurred while reading pseudo element font usage')
+        font_usage += get_pseudo_element_font_usage(pseudo_element_font_usage, self.first_letter_pat, self.parser)
         exclude = {'\n', '\r', '\t'}
         self.font_usage_map[self.container.abspath_to_name(self.current_item)] = fu = defaultdict(dict)
         bad_fonts = {'serif', 'sans-serif', 'monospace', 'cursive', 'fantasy', 'sansserif', 'inherit'}
@@ -314,11 +354,11 @@ class StatsCollector(object):
             if not isinstance(font_families, dict):
                 raise Exception('Unknown error occurred while reading font families')
             self.font_spec_map[self.container.abspath_to_name(self.current_item)] = fs = set()
+            for font_dict, text, pseudo in pseudo_element_font_usage:
+                font_families[font_dict['font-family']] = True
             for raw in font_families.iterkeys():
-                style = self.parser.parseStyle('font-family:' + raw, validate=False).getProperty('font-family')
+                for x in parse_font_families(self.parser, raw):
-                for x in style.propertyValue:
+                    if x.lower() not in bad_fonts:
-                    x = x.value
-                    if x and x.lower() not in bad_fonts:
                         fs.add(x)
 
 if __name__ == '__main__':
@@ -327,5 +367,3 @@ if __name__ == '__main__':
     default_log.filter_level = default_log.DEBUG
     ebook = get_container(sys.argv[-1], default_log)
     print (StatsCollector(ebook, do_embed=True).font_stats)
-
-