Skip to main content

Documentation Index

Fetch the complete documentation index at: https://ahvn.top/llms.txt

Use this file to discover all available pages before exploring further.

String utilities provide compact display helpers plus Markdown builders used by prompts, reports, generated docs, and tests.

​
1. Keep display text short

Use truncate or shorten when CLI output, logs, or reports should show a compact preview. 
from heavenbase.utils.strings import shorten

preview = shorten("a very long explanation that should not dominate a table", cutoff=24)
The helper returns the original string when cutoff is negative or the value already fits.

​
2. Format blocks

Use indent and dedent for generated text blocks. They wrap Python’s text formatting behavior behind one project-level import. 
from heavenbase.utils.strings import dedent, indent

body = dedent("""
    First line
    Second line
""")

print(indent(body, tab=2))
Use md_block, code_block, file_block, and json_block for fenced Markdown output. Fences expand automatically when the content already contains backticks. 
from heavenbase.utils.strings import code_block, json_block

print(code_block("print('hello')", lang="python"))
print(json_block({"status": "ok"}))
Use bullet_list, numbered_list, and bullet_dict for compact lists that preserve indentation for multi-line values. 
from heavenbase.utils.strings import bullet_dict

print(bullet_dict({"backend": "sqlite", "rows": 1000}))

​
3. Build Markdown sections

Use md_section to build nested Markdown with stable blank lines and heading levels. 
from heavenbase.utils.strings import bullet_list, code_block, md_section

report = md_section(
    title="Run report",
    content="Summary text.",
    sections=[
        {"title": "Inputs", "content": bullet_list(["sqlite", "1000 rows"])},
        {"title": "Snippet", "content": code_block("result = hb.open('demo')", lang="python")},
    ],
    end="Done.",
)
The rendered section uses exactly one blank line between headings, body blocks, nested sections, and the ending paragraph. That makes generated Markdown readable without extra post-processing.

​
4. Normalize names and text

Use snake when free-form names must become simple identifiers for diagnostics or generated artifacts. 
from heavenbase.utils.strings import snake

assert snake("Search Result") == "search_result"
For entity IDs and backend-safe names, use the dedicated validation helpers in heavenbase.utils: check_entity_id, check_workspace_id, entity_table_name, and workspace_index_name. For lightweight matching, is_delimiter, generate_ngrams, resolve_match_conflicts, and asymmetric_jaccard_score are available. normalize_text uses spaCy when en_core_web_sm is installed and raises a clear error otherwise.

​
Further Exploration

Related resources:
  • Common utilities - typing helpers, debug output, colors, and hashing.
  • Schema - entity and field naming conventions.