From b1dbb3d40ce29526e81c842e3826eb2091458be8 Mon Sep 17 00:00:00 2001
From: "Peter J. Holzer" <hjp@hjp.at>
Date: Mon, 20 May 2019 00:32:33 +0200
Subject: [PATCH] Implement format=flowed and fix charset handling for
 text/plain

---
 mbox2web                              | 105 ++++++++++++++++++++++++--
 style/debug.css                       |  20 +++++
 templates/body_text_plain_flowed.html |   8 ++
 3 files changed, 128 insertions(+), 5 deletions(-)
 create mode 100644 templates/body_text_plain_flowed.html

diff --git a/mbox2web b/mbox2web
index 7f03952..cd3e842 100755
--- a/mbox2web
+++ b/mbox2web
@@ -130,11 +130,25 @@ partial_message_cache = {}
 
 def render_body(msg, extra=None):
     def render_text_plain(msg, extra=None):
-        bodytmpl = jenv.get_template("body_text_plain.html")
-        context = {
-            "body": msg.get_payload(decode=True).decode(msg.get_charset() or "iso-8859-1")
-        }
-        return bodytmpl.render(context)
+        # msg.get_charset() doesn't work
+        ct_params = dict(msg.get_params() or [])
+        charset = ct_params.get("charset", "iso-8859-1")
+        format = ct_params.get("format", "fixed")
+        if format == "fixed":
+            bodytmpl = jenv.get_template("body_text_plain.html")
+            context = {
+                "body": msg.get_payload(decode=True).decode(charset)
+            }
+            return bodytmpl.render(context)
+        elif format == "flowed":
+            bodytmpl = jenv.get_template("body_text_plain_flowed.html")
+            parthtml = TextFlowedPart(msg).as_string()
+            context = {
+                "body": jinja2.Markup(parthtml),
+            }
+            return bodytmpl.render(context)
+        else:
+            raise NotImplementedError()
     def render_multipart_mixed(msg, extra=None):
         parts = msg.get_payload()
         if type(parts) == str:
@@ -627,6 +641,87 @@ class TextEnrichedPart:
         return self.stack[0].as_string()
 
 
+class TextFlowedPart:
+    def __init__(self, msg):
+        self.quote_depth = 0
+        self.current_line = ""
+        self.flowed = False
+        self.lines = []
+        self.buffer_filled = False
+
+        ct_params = dict(msg.get_params())
+        charset = ct_params.get("charset", "iso-8859-1")
+        format = ct_params.get("format", "fixed")
+        delsp = ct_params.get("delsp", "no") == "yes"
+        raw_text = msg.get_payload(decode=True).decode(charset)
+        raw_lines = raw_text.split("\n")
+
+        for rl in raw_lines:
+            quote_depth = 0
+            while rl[:1] == ">":
+                quote_depth += 1
+                rl = rl[1:]
+            if rl[:1] == " ":
+                rl = rl[1:]
+            if rl == "-- ":
+                flowed = None
+            elif rl[-1:] == " ":
+                flowed = True
+                if delsp:
+                    rl = rl[:-1]
+            else:
+                flowed = False
+            self.add_buffer(rl, quote_depth, flowed)
+        self.flush()
+
+    def add_buffer(self, line, quote_depth, flowed):
+        if flowed is None:
+            self.flush()
+            flowed = False
+        if quote_depth != self.quote_depth:
+            self.flush()
+            self.quote_depth = quote_depth
+        self.current_line += line
+        self.flowed |= flowed
+        self.buffer_filled = True
+        if not flowed:
+            self.flush()
+
+    def flush(self):
+        if self.buffer_filled:
+            self.lines.append({
+                            "quote_depth": self.quote_depth,
+                            "flowed": self.flowed,
+                            "content": self.current_line
+                        })
+        self.current_line = ""
+        self.flowed = False
+        self.buffer_filled = False
+
+    def as_string(self):
+        prev_quote_depth = 0
+        s = ""
+        for ln in self.lines:
+            while ln["quote_depth"] > prev_quote_depth:
+                s += "<blockquote>\n"
+                prev_quote_depth += 1
+            while ln["quote_depth"] < prev_quote_depth:
+                s += "</blockquote>\n"
+                prev_quote_depth -= 1
+            if ln["flowed"]:
+                s += "<p class='flowed'>" + html.escape(ln["content"]) + "</p>\n"
+            else:
+                s += "<p class='fixed'>" + html.escape(ln["content"]) + "</p>\n"
+        while 0 < prev_quote_depth:
+            s += "</blockquote>"
+            prev_quote_depth -= 1
+        return s
+
+
+
+
+
+
 for f in sys.argv[1:]:
     print("F", f, file=sys.stderr)
     mb = mailbox.mbox(f)
diff --git a/style/debug.css b/style/debug.css
index ccec653..acd6423 100644
--- a/style/debug.css
+++ b/style/debug.css
@@ -42,3 +42,23 @@ th {
 .partinner.html {
     max-width: 40rem;
 }
+
+p.fixed {
+    font-family: monospace;
+    white-space: pre;
+    margin: 0.2em;
+}
+
+p.flowed {
+    font-family: monospace;
+    white-space: pre-wrap;
+    max-width: 60em;
+    margin: 0.2em;
+}
+
+.text_plain_flowed blockquote {
+    border-left: solid 0.2em #004;
+    padding-left: 0.8em;
+    margin: 0em;
+    background: #EEF;
+}
diff --git a/templates/body_text_plain_flowed.html b/templates/body_text_plain_flowed.html
new file mode 100644
index 0000000..f774190
--- /dev/null
+++ b/templates/body_text_plain_flowed.html
@@ -0,0 +1,8 @@
+<div class="partouter">
+    <div class="partheader">
+       text/plain
+    </div>
+    <div class="partinner text_plain_flowed">
+{{body}}
+    </div>
+</div>