misc: Added initial codebase.

htmlparser.py

@@ -0,0 +1,187 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# System packages
+import pathlib
+from datetime import datetime
+from html.parser import HTMLParser
+from typing import List, Tuple, Optional, Set
+from enum import IntEnum
+
+path_project = pathlib.Path("/home/gmartin/Workspace/DiscordPrettyPrinter")
+
+
+class DiscordRollTok(IntEnum):
+    """
+    List the different times discords tokens has been modified.
+    """
+    ED_20230709 = 1  # Exports réalisés le 2023-07-09
+    ED_20240605 = 2  # Exports réalisés le 2024-06-05
+
+# Tokens used to parse pseudos.
+PSEUDO_TAG: dict[DiscordRollTok, str] = {
+    DiscordRollTok.ED_20230709: "username-h_Y3Us",
+    DiscordRollTok.ED_20240605: "username__0b0e7",
+}
+# Tokens used to parse reaction counts.
+REACTCOUNT_TAG: dict[DiscordRollTok, str] = {
+    DiscordRollTok.ED_20230709: "reactionCount-26U4As",
+    DiscordRollTok.ED_20240605: "reactionCount__2c34d",
+}
+# Tokens used to parse the divider and the date.
+DIVIDER_TAG: dict[DiscordRollTok, str] = {
+    DiscordRollTok.ED_20230709: "divider-IqmEqJ",
+    DiscordRollTok.ED_20240605: "divider__01aed",
+}
+
+
+def check_tagattr(attrs: List[Tuple[str, str]], tag: str) -> bool:
+    """Checks if the given tag, e.g. 'username-h_Y3Us', is inside the list of attributes values.
+    """
+    attrs_keyvalues: List[str] = list(zip(*attrs))
+    if len(attrs_keyvalues):
+        for attr_value in attrs_keyvalues[1]:
+            if tag in attr_value:
+                return True
+    return False
+
+
+class DiscordHTMLParser(HTMLParser):
+
+    def __init__(self, keep_time: bool, convert_charrefs=True):
+        super().__init__(convert_charrefs=convert_charrefs)
+
+        self.keep_time = keep_time
+
+        # Parsed text to be exported.
+        self.export_txt: str = ""
+        self.nb_messages: int = 0
+        self.nb_duplicate: int = 0
+
+        # Information about what is being parsed.
+        self.is_scan_message: bool = False
+        self.is_scan_pseudo: bool = False
+        self.is_scan_time: bool = False
+        self.is_scan_reactioncount: bool = False
+        self.is_scan_datedivider: bool = False
+
+        # When parsing a time element, keep its datetime to print it in the message.
+        self.current_dt: Optional[datetime] = None
+
+        # Keep track of which message and date separator have been added.
+        self.set_idmessage: Set[str] = set()
+        self.set_dateseparator: Set[str] = set()
+
+    def handle_starttag(self, tag, attrs):
+
+        # Add to the export the tags used to format the text.
+        match tag:
+            case "em":
+                if self.is_scan_message:
+                    self.export_txt += "<em>"
+            case "strong":
+                if self.is_scan_message:
+                    self.export_txt += "<b>"
+            case "li":
+                if check_tagattr(attrs, "chat-messages"):
+
+                    # Retrieve the id of the message.
+                    id_message: str = ""
+                    for key, value in attrs:
+                        if "chat-messages" in value:
+                            id_message = value.split("chat-messages-")[1]
+
+                    # If the message has already been seen, don't keep it.
+                    if id_message in self.set_idmessage:
+                        self.nb_duplicate += 1
+                    else:
+                        self.set_idmessage.add(id_message)
+
+                        self.nb_messages += 1
+                        self.is_scan_message = True
+                        self.export_txt += f"\n<li class=\"message\" id_message=\"{id_message}\">"
+            case "span":
+                # Tell whether the pseudo is being scanned or not.
+                if check_tagattr(attrs, PSEUDO_TAG[DiscordRollTok.ED_20240605]):
+                    self.is_scan_pseudo = True
+            case "div":
+                # Tell whether the reaction count is being scanned or not.
+                if check_tagattr(attrs, REACTCOUNT_TAG[DiscordRollTok.ED_20240605]):
+                    self.is_scan_reactioncount = True
+                elif check_tagattr(attrs, DIVIDER_TAG[DiscordRollTok.ED_20240605]):
+
+                    # Retrieve which date separator is being scanned.
+                    date_separator: str = ""
+                    for key, value in attrs:
+                        if key == "aria-label":
+                            date_separator = value
+
+                    if date_separator not in self.set_dateseparator:
+                        self.set_dateseparator.add(date_separator)
+                        self.is_scan_datedivider = True
+            case "time":
+                if self.is_scan_message:
+                    if not check_tagattr(attrs, "Edited"):
+                        self.is_scan_time = True
+                        for key, value in attrs:
+                            if key == "datetime":
+                                self.current_dt = datetime.strptime(value, "%Y-%m-%dT%H:%M:%S.%fZ")
+
+    def handle_endtag(self, tag):
+
+        # Add to the export the tags used to format the text.
+        match tag:
+            case "em":
+                if self.is_scan_message:
+                    self.export_txt += "</em>"
+            case "strong":
+                if self.is_scan_message:
+                    self.export_txt += "</b>"
+            case "li":
+                if self.is_scan_message:
+                    self.is_scan_message = False
+                    self.export_txt += "</li>"
+            case "span":
+                self.is_scan_pseudo = False
+            case "div":
+                self.is_scan_reactioncount = False
+                self.is_scan_datedivider = False
+            case "time":
+                self.is_scan_time = False
+                self.current_dt = None
+
+    def handle_data(self, data):
+
+        if self.is_scan_message:
+            if self.is_scan_reactioncount:
+                return
+            elif self.is_scan_time and self.keep_time:
+                if self.current_dt is not None:
+                    self.export_txt += f"<span class=\"time\">[{self.current_dt.strftime('%Y-%m-%d %H:%M')}]</span> "
+                    self.current_dt = None
+            elif self.is_scan_pseudo:
+                pseudo_ansi = ''.join(i for i in data.lower() if ord(i) < 128)
+                self.export_txt += f"<span class=\"pseudo_{pseudo_ansi}\"><b>"
+                self.export_txt += data
+                self.export_txt += "</b></span>"
+            elif not (data in [""]):
+                data = data.replace("\n", "\n<br/>")
+                self.export_txt += data
+        elif self.is_scan_datedivider:
+            self.export_txt += f"<div class=\"dateSeparator\">{data}</div>\n"
+
+    def export_messages(self) -> str:
+        before_export: str = ""
+        after_export: str = ""
+
+        before_export += "<html>\n"
+        before_export += "<head>\n"
+        before_export += "<link rel=\"stylesheet\" type=\"text/css\" href=\"style.css\" />\n"
+        before_export += "</head>\n"
+        before_export += "<body>\n"
+        before_export += "<ul>\n"
+
+        after_export += "</ul>\n"
+        after_export += "</body>\n"
+        after_export += "</html>\n"
+        return before_export + self.export_txt + after_export

main.py

@@ -0,0 +1,37 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# System packages
+import pathlib
+from pathlib import Path
+
+# Project package
+from htmlparser import DiscordHTMLParser
+
+path_project = pathlib.Path("/home/gmartin/Workspace/DiscordPrettyPrinter")
+
+if __name__ == '__main__':
+    parser = DiscordHTMLParser(keep_time=False)
+    folder_name = "sar"
+    file_name = "mp"
+
+    xml_path: pathlib.Path = path_project / "input" / folder_name / (file_name + ".xml")
+    
+    with open(xml_path, "r") as file_xml:
+        xml_str = ""
+        for line in file_xml:
+            xml_str += line
+        parser.feed(xml_str)
+        # print(parser.export_messages())
+        print(f"Parsed: {xml_path}")
+        print(f"{parser.nb_messages} messages found (and {parser.nb_duplicate} duplicates).")
+
+        # Creation of the output folder if it does not exist.
+        path_file_output: Path = (
+            path_project / "output" / folder_name / (file_name + ".html")
+        )
+        path_file_output.parent.mkdir(parents=True, exist_ok=True)
+
+        # Export of the parsed text and formatted into html.
+        with open(path_file_output, "w") as file_htmloutput:
+            file_htmloutput.write(parser.export_messages())