From d14f0200d954e16b4a09b861baa053ec6822abfe Mon Sep 17 00:00:00 2001
From: Jan Hartig <jan.hartig@ptb.de>
Date: Tue, 15 Apr 2025 17:54:54 +0200
Subject: [PATCH] mailservice: Add transcript support
---
fix_transcript.py | 29 ------------------
localisations.toml | 8 +++++
mailservice.py | 76 ++++++++++++++++++++++++++++++++++++----------
3 files changed, 68 insertions(+), 45 deletions(-)
delete mode 100644 fix_transcript.py
diff --git a/fix_transcript.py b/fix_transcript.py
deleted file mode 100644
index 1848391..0000000
--- a/fix_transcript.py
+++ /dev/null
@@ -1,29 +0,0 @@
-import re
-
-transcript_pattern = re.compile(r"^\[(\w+)]: (.+)$")
-
-with open(r"transcript.txt", encoding="utf-8") as f:
- transcript = f.read()
-
-lines = []
-for line in transcript.split("\n"):
- match = re.fullmatch(transcript_pattern, line)
-
- if match:
- lines.append(match.groups())
-
-output = ""
-first_line = True
-previous_speaker = None
-for speaker, line in lines:
- if previous_speaker == speaker:
- output += f" {line}"
- else:
- previous_speaker = speaker
- output += f"{'\n' if not first_line else ''}{speaker}: {line}"
- first_line = False
-
-print(output)
-
-with open(r"transcript_fixed.txt", "w", encoding="utf-8") as f:
- f.write(output)
diff --git a/localisations.toml b/localisations.toml
index e9f1787..8874bf5 100644
--- a/localisations.toml
+++ b/localisations.toml
@@ -134,3 +134,11 @@ en = "Create transcript instead?"
[ footer_transcript ]
de = "Stattdessen Untertitel erstellen?"
en = "Create subtitles instead?"
+
+[ mail.transcribe_subject ]
+de = "Ihr Transkript"
+en = "Your transcript"
+
+[ mail.transcribe_content ]
+de = "Das Transkript für Ihre Datei '{}' wurden erstellt.\nSie finden das Resultat im Anhang dieser E-Mail."
+en = "The transcript for your file '{}' have been created.\nThe result is attached to this email."
\ No newline at end of file
diff --git a/mailservice.py b/mailservice.py
index bfe1e67..7627a78 100644
--- a/mailservice.py
+++ b/mailservice.py
@@ -1,3 +1,4 @@
+import io
import json
import shutil
import smtplib
@@ -5,6 +6,7 @@ import tomllib
from email.message import EmailMessage
from os import scandir
from pathlib import Path
+import re
import requests
from requests.auth import HTTPBasicAuth
@@ -67,26 +69,66 @@ def main(end):
language = metadata["language"]
msg = EmailMessage()
- msg["Subject"] = localisations["mail"]["subject"][language]
- msg["From"] = config["MAIL"]["FROM"]
- msg["To"] = metadata["email"]
- msg.set_content(
- localisations["mail"]["content"][language].format(metadata["filename"])
- )
+ if metadata["job_type"] == "subtitle":
+ msg["Subject"] = localisations["mail"]["subject"][language]
+ msg["From"] = config["MAIL"]["FROM"]
+ msg["To"] = metadata["email"]
- # filename.language.vtt
- if metadata["video_language"] == "None":
- filename = Path(metadata["filename"]).with_suffix(".vtt").name
- else:
- filename = (
- Path(metadata["filename"])
- .with_suffix(".{}.vtt".format(metadata["video_language"]))
- .name
+ msg.set_content(
+ localisations["mail"]["content"][language].format(metadata["filename"])
)
- with open(Path(job).joinpath("audio.vtt")) as f:
- msg.add_attachment(f.read(), filename=filename)
+ # filename.language.vtt
+ if metadata["video_language"] == "None":
+ filename = Path(metadata["filename"]).with_suffix(".vtt").name
+ else:
+ filename = (
+ Path(metadata["filename"])
+ .with_suffix(".{}.vtt".format(metadata["video_language"]))
+ .name
+ )
+
+ with open(Path(job).joinpath("audio.vtt")) as f:
+ msg.add_attachment(f.read(), filename=filename)
+
+ elif metadata["job_type"] == "transcript":
+ msg["Subject"] = localisations["mail"]["transcribe_subject"][language]
+ msg["From"] = config["MAIL"]["FROM"]
+ msg["To"] = metadata["email"]
+
+ msg.set_content(
+ localisations["mail"]["transcribe_content"][language].format(
+ metadata["filename"]
+ )
+ )
+
+ # filename.txt
+ filename = Path(metadata["filename"]).with_suffix(".txt").name
+
+ with open(Path(job).joinpath("audio.txt"), encoding="utf-8") as f:
+ transcript = f.read()
+
+ # reformat transcript
+ lines = []
+ for line in transcript.split("\n"):
+ match = re.fullmatch(transcript_pattern, line)
+
+ if match:
+ lines.append(match.groups())
+
+ transcript = ""
+ first_line = True
+ previous_speaker = None
+ for speaker, line in lines:
+ if previous_speaker == speaker:
+ transcript += f" {line}"
+ else:
+ previous_speaker = speaker
+ transcript += f"{'\n' if not first_line else ''}{speaker}: {line}"
+ first_line = False
+
+ msg.add_attachment(transcript, filename=filename)
s.send_message(msg)
@@ -146,6 +188,8 @@ if __name__ == "__main__":
end = Event()
+ transcript_pattern = re.compile(r"^\[(\w+)]: (.+)$")
+
def handler(signum, frame):
global end
print(signum)
--
GitLab