Saving & Export
Requires pip install "unihra[report]" (pandas + openpyxl).
Excel Report
python
client.save_report(result, "report.xlsx")Generates a multi-sheet workbook:
| Sheet | Content |
|---|---|
| Anchors | Anchor text analysis with frequency and href links |
| Parsed Pages | URLs included in the analysis |
| Failed Pages | Requested URLs not included in the analysis |
| Page Structure | Meta titles, H1, headings, content volume |
| Umbrella Gaps | Umbrella analysis — lemmas to add with recommendations |
| Word Analysis | TF-IDF words with action labels |
| N-Grams | Phrase patterns |
| Triplets | Knowledge graph entities and facts (if triplet_analysis=True) |
| Triplets Gaps | Missing topics by severity (if triplet_analysis=True) |
Color coding is applied automatically:
- 🔴 Red — word/anchor absent from your page
- 🟢 Green — word/anchor present on your page
- 🟠 Amber — important / main tier in triplets
Disable Styling
python
client.save_report(result, "report.xlsx", style_output=False)CSV Report
Exports only block_comparison (word analysis) to CSV:
python
client.save_report(result, "report.csv")Pandas DataFrame
python
# Requires pandas: pip install "unihra[report]"
df_gaps = client.get_dataframe(result, "umbrella_analysis")
df_words = client.get_dataframe(result, "block_comparison")
df_ngrams = client.get_dataframe(result, "ngrams_analysis")
df_anchors = client.get_dataframe(result, "anchors_analysis")
df_parsed = client.get_dataframe(result, "parsed_pages")
df_failed = client.get_dataframe(result, "failed_pages")
df_struct = client.get_dataframe(result, "page_structure")
# Triplets (extended mode only)
df_entities = client.get_dataframe(result, "triplets_analysis")
df_gaps_t = client.get_dataframe(result, "triplets_gaps")Backward compatibility
umbrella_analysis replaces the old semantic_context_analysis key. Both are accepted by get_dataframe().
Save Raw JSON (Segmented)
Splits the result into separate JSON files per section:
python
manifest = client.analyze_and_save(
own_page="https://example.com",
competitors=["https://comp.com"],
)
# {
# "analysis_id": "550e8400-...",
# "files": {
# "gaps": "/path/to/550e8400-.../gaps.json",
# "words": "/path/to/.../words.json",
# "ngrams": "/path/to/.../ngrams.json",
# "anchors": "/path/to/.../anchors.json",
# "triplets": "/path/to/.../triplets.json",
# "structure": "/path/to/.../structure.json"
# }
# }
print(manifest["files"]["gaps"])Files are saved to ./unihra_results/{task_id}/ by default. Change it:
python
client = UnihraClient(api_key="YOUR_KEY", storage_dir="/data/seo_results")