Make scripts better

This commit is contained in:
Слонова Анна 2026-04-30 10:57:18 +03:00
parent ba56147e95
commit e8ad7df469
12 changed files with 614 additions and 432 deletions

View file

@ -1,60 +1,73 @@
#!/bin/bash
# generate_report.sh — Full pipeline for generating meeting report (without diagrams)
# Usage: ./generate_report.sh /absolute/path/to/meeting_folder
# Example: ./generate_report.sh /app/hermes_data/meetings/2026-04-15
# generate_report.sh — Simplified pipeline: transcription + merge + PDF
# Usage: ./generate_report.sh <meeting-date-dir>
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
# Load .env if exists (for hotwords etc.)
if [ -f "$SCRIPT_DIR/.env" ]; then
set -a
source "$SCRIPT_DIR/.env"
set +a
fi
if [ $# -lt 1 ]; then
echo "Usage: $0 <absolute_path_to_meeting_folder>"
echo "Example: $0 /app/hermes_data/meetings/2026-04-15"
echo "Usage: $0 <meeting-date-dir>"
exit 1
fi
MEETING_DIR="$1"
# If relative path provided, convert to absolute
if [[ "$MEETING_DIR" != /* ]]; then
MEETING_DIR="$SCRIPT_DIR/$MEETING_DIR"
fi
# Resolve absolute path
MEETING_DIR="$(realpath "$MEETING_DIR")"
MEETING_DIR="$SCRIPT_DIR/$1"
TRANSCRIPTION_DIR="$MEETING_DIR/transcription"
DIAGRAMS_DIR="$MEETING_DIR/diagrams"
if [ ! -d "$MEETING_DIR" ]; then
echo "Error: Meeting directory not found: $MEETING_DIR"
exit 1
fi
# ============================================================
# ------------------------------------------------------------
# Step 1: Transcription (skip if already done)
# ============================================================
if [ -d "$MEETING_DIR/transcription" ] && [ -f "$MEETING_DIR/transcription/plain_text.txt" ]; then
echo "[1/2] Transcription already exists, skipping."
# ------------------------------------------------------------
if [ -d "$TRANSCRIPTION_DIR" ] && [ -f "$TRANSCRIPTION_DIR/plain_text.txt" ]; then
echo "[1/3] Transcription already exists, skipping."
else
echo "[1/2] Running transcription..."
echo "[1/3] Running transcription..."
bash "$SCRIPT_DIR/transcribe.sh" "$MEETING_DIR"
fi
# ============================================================
# Step 2: Generate PDF from markdown
# ============================================================
echo "[2/2] Generating PDF..."
# ------------------------------------------------------------
# Step 2: Merge transcriptions (if both saramonic and h2n exist)
# ------------------------------------------------------------
echo "[2/3] Merging transcriptions where possible..."
if [ -f "$TRANSCRIPTION_DIR/saramonic.json" ] && [ -f "$TRANSCRIPTION_DIR/h2n_xy.json" ]; then
echo " Merging saramonic (primary) + h2n_xy (secondary) → merged.json"
python3 "$SCRIPT_DIR/merge_transcriptions.py" \
"$TRANSCRIPTION_DIR/saramonic.json" \
"$TRANSCRIPTION_DIR/h2n_xy.json" \
"$MEETING_DIR"
elif [ -f "$TRANSCRIPTION_DIR/saramonic.json" ] && [ -f "$TRANSCRIPTION_DIR/h2n_ms.json" ]; then
echo " Merging saramonic (primary) + h2n_ms (secondary) → merged.json"
python3 "$SCRIPT_DIR/merge_transcriptions.py" \
"$TRANSCRIPTION_DIR/saramonic.json" \
"$TRANSCRIPTION_DIR/h2n_ms.json" \
"$MEETING_DIR"
else
echo " Only one transcription source found or none — copying plain text to merged_plain.txt"
# Try to find any existing plain text
PLAIN_SRC=$(find "$TRANSCRIPTION_DIR" -name "*_plain.txt" | head -1)
if [ -n "$PLAIN_SRC" ]; then
cp "$PLAIN_SRC" "$MEETING_DIR/merged_plain.txt"
echo " Copied $PLAIN_SRC -> merged_plain.txt"
else
echo " No plain text found yet — will be created after transcription finishes."
fi
fi
# ------------------------------------------------------------
# Step 3: Generate PDF from report.md (if exists)
# ------------------------------------------------------------
echo "[3/3] Generating PDF from report.md (if present)..."
REPORT_MD="$MEETING_DIR/report.md"
REPORT_PDF="$MEETING_DIR/report.pdf"
if [ ! -f "$REPORT_MD" ]; then
echo " Error: report.md not found at $REPORT_MD"
exit 1
echo " No report.md found. Agent must write this file first."
echo " After writing the report, run: pandoc $REPORT_MD -o $REPORT_PDF ..."
exit 0
fi
cd "$MEETING_DIR"