From f9563cddcdb33432c8c5445e7f43108becba3728 Mon Sep 17 00:00:00 2001
From: tastytea <tastytea@tastytea.de>
Date: Fri, 17 May 2019 05:43:17 +0200
Subject: [PATCH] Replaced regular expressions in with find & replace, where
 possible.

---
 CMakeLists.txt |  2 +-
 src/uri.cpp    | 50 +++++++++++++++++++++++++++++++++++++-------------
 src/uri.hpp    |  3 ++-
 3 files changed, 40 insertions(+), 15 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 6f5e72b..2973673 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,6 +1,6 @@
 cmake_minimum_required (VERSION 3.2)
 project(remwharead
-  VERSION 0.1.1
+  VERSION 0.1.2
   LANGUAGES CXX
 )
 
diff --git a/src/uri.cpp b/src/uri.cpp
index 71eea6c..db7d5c1 100644
--- a/src/uri.cpp
+++ b/src/uri.cpp
@@ -98,29 +98,53 @@ const string URI::strip_html(const string &html)
 {
     string out;
 
-    out = regex_replace(html, regex("<script[^<]+"), ""); // Remove JavaScript.
-    out = regex_replace(out, regex("<style[^<]+"), "");   // Remove CSS.
-    out = remove_html_tags(out);                          // Remove tags.
-    out = regex_replace(out, regex("\r"), "");            // Remove CR.
+    out = remove_html_tags(html, "script") // Remove JavaScript.
+    out = remove_html_tags(out, "style");  // Remove CSS.
+    out = remove_html_tags(out);           // Remove tags.
+
+    size_t pos = 0;
+    while ((pos = out.find("\r")) != std::string::npos)   // Remove CR.
+    {
+        out.replace(pos, 1, "");
+    }
+
     out = regex_replace(out, regex("\\s+\n"), "\n"); // Remove trailing space.
     out = regex_replace(out, regex("\n{2,}"), "\n"); // Reduce newlines.
 
     return unescape_html(out);
 }
-const string URI::remove_html_tags(const string &html)
+const string URI::remove_html_tags(const string &html, const string &tag)
 {
     // NOTE: I did this with regex_replace before, but libstdc++ segfaulted.
     string out;
-    size_t pos = 0;
-    while (pos != std::string::npos)
+    if (tag.empty())
     {
-        size_t startpos = html.find('<', pos);
-        size_t endpos = html.find('>', startpos);
-        out += html.substr(pos, startpos - pos);
-        pos = endpos;
-        if (pos != std::string::npos)
+        size_t pos = 0;
+        while (pos != std::string::npos)
         {
-            ++pos;
+            size_t startpos = html.find('<', pos);
+            size_t endpos = html.find('>', startpos);
+            out += html.substr(pos, startpos - pos);
+            pos = endpos;
+            if (pos != std::string::npos)
+            {
+                ++pos;
+            }
+        }
+    }
+    else
+    {
+        size_t pos = 0;
+        out = html;
+        while ((pos = out.find("<" + tag)) != std::string::npos)
+        {
+            size_t endpos = out.find("</" + tag, pos);
+            if (endpos == std::string::npos)
+            {
+                break;
+            }
+            endpos += 3 + tag.length(); // tag + </ + >
+            out.replace(pos, endpos - pos, "");
         }
     }
 
diff --git a/src/uri.hpp b/src/uri.hpp
index 04caec6..3217013 100644
--- a/src/uri.hpp
+++ b/src/uri.hpp
@@ -44,7 +44,8 @@ protected:
     const string extract_title(const string &html);
     const string extract_description(const string &html);
     const string strip_html(const string &html);
-    const string remove_html_tags(const string &html);
+    //! Remove all HTML tags. If tag is not empty, remove tag and its content.
+    const string remove_html_tags(const string &html, const string &tag = "");
     const string unescape_html(const string &html);
     const string remove_newlines(const string &text);
 };