feat(utils): add abbreviate_path for smart path/URL truncation

This commit is contained in:
chengyongru 2026-04-07 11:31:00 +08:00 committed by Xubin Ren
parent 02597c3ec9
commit f452af6c62
2 changed files with 185 additions and 0 deletions

106
nanobot/utils/path.py Normal file
View File

@ -0,0 +1,106 @@
"""Path abbreviation utilities for display."""
from __future__ import annotations
import os
import re
from urllib.parse import urlparse
def abbreviate_path(path: str, max_len: int = 40) -> str:
"""Abbreviate a file path or URL, preserving basename and key directories.
Strategy:
1. Return as-is if short enough
2. Replace home directory with ~/
3. From right, keep basename + parent dirs until budget exhausted
4. Prefix with /
"""
if not path:
return path
# Handle URLs: preserve scheme://domain + filename
if re.match(r"https?://", path):
return _abbreviate_url(path, max_len)
# Normalize separators to /
normalized = path.replace("\\", "/")
# Replace home directory
home = os.path.expanduser("~").replace("\\", "/")
if normalized.startswith(home + "/"):
normalized = "~" + normalized[len(home):]
elif normalized == home:
normalized = "~"
# Return early only after normalization and home replacement
if len(normalized) <= max_len:
return normalized
# Split into segments
parts = normalized.rstrip("/").split("/")
if len(parts) <= 1:
return normalized[:max_len - 1] + "\u2026"
# Always keep the basename
basename = parts[-1]
# Budget: max_len minus "…/" prefix (2 chars) minus "/" separator minus basename
budget = max_len - len(basename) - 3 # -3 for "…/" + final "/"
# Walk backwards from parent, collecting segments
kept: list[str] = []
for seg in reversed(parts[:-1]):
needed = len(seg) + 1 # segment + "/"
if not kept and needed <= budget:
kept.append(seg)
budget -= needed
elif kept:
needed_with_sep = len(seg) + 1
if needed_with_sep <= budget:
kept.append(seg)
budget -= needed_with_sep
else:
break
else:
break
kept.reverse()
if kept:
return "\u2026/" + "/".join(kept) + "/" + basename
return "\u2026/" + basename
def _abbreviate_url(url: str, max_len: int = 40) -> str:
"""Abbreviate a URL keeping domain and filename."""
if len(url) <= max_len:
return url
parsed = urlparse(url)
domain = parsed.netloc # e.g. "example.com"
path_part = parsed.path # e.g. "/api/v2/resource.json"
# Extract filename from path
segments = path_part.rstrip("/").split("/")
basename = segments[-1] if segments else ""
if not basename:
# No filename, truncate URL
return url[: max_len - 1] + "\u2026"
budget = max_len - len(domain) - len(basename) - 4 # "…/" + "/"
if budget < 0:
return domain + "/\u2026" + basename[: max_len - len(domain) - 4]
# Build abbreviated path
kept: list[str] = []
for seg in reversed(segments[:-1]):
if len(seg) + 1 <= budget:
kept.append(seg)
budget -= len(seg) + 1
else:
break
kept.reverse()
if kept:
return domain + "/\u2026/" + "/".join(kept) + "/" + basename
return domain + "/\u2026/" + basename

View File

@ -0,0 +1,79 @@
"""Tests for abbreviate_path utility."""
import os
from nanobot.utils.path import abbreviate_path
class TestAbbreviatePathShort:
def test_short_path_unchanged(self):
assert abbreviate_path("/home/user/file.py") == "/home/user/file.py"
def test_exact_max_len_unchanged(self):
path = "/a/b/c" # 7 chars
assert abbreviate_path("/a/b/c", max_len=7) == "/a/b/c"
def test_basename_only(self):
assert abbreviate_path("file.py") == "file.py"
def test_empty_string(self):
assert abbreviate_path("") == ""
class TestAbbreviatePathHome:
def test_home_replacement(self):
home = os.path.expanduser("~")
result = abbreviate_path(f"{home}/project/file.py")
assert result.startswith("~/")
assert result.endswith("file.py")
def test_home_preserves_short_path(self):
home = os.path.expanduser("~")
result = abbreviate_path(f"{home}/a.py")
assert result == "~/a.py"
class TestAbbreviatePathLong:
def test_long_path_keeps_basename(self):
path = "/a/b/c/d/e/f/g/h/very_long_filename.py"
result = abbreviate_path(path, max_len=30)
assert result.endswith("very_long_filename.py")
assert "\u2026" in result
def test_long_path_keeps_parent_dir(self):
path = "/a/b/c/d/e/f/g/h/src/loop.py"
result = abbreviate_path(path, max_len=30)
assert "loop.py" in result
assert "src" in result
def test_very_long_path_just_basename(self):
path = "/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/file.py"
result = abbreviate_path(path, max_len=20)
assert result.endswith("file.py")
assert len(result) <= 20
class TestAbbreviatePathWindows:
def test_windows_drive_path(self):
path = "D:\\Documents\\GitHub\\nanobot\\src\\utils\\helpers.py"
result = abbreviate_path(path, max_len=40)
assert result.endswith("helpers.py")
assert "nanobot" in result
def test_windows_home(self):
home = os.path.expanduser("~")
path = os.path.join(home, ".nanobot", "workspace", "log.txt")
result = abbreviate_path(path)
assert result.startswith("~/")
assert "log.txt" in result
class TestAbbreviatePathURLs:
def test_url_keeps_domain_and_filename(self):
url = "https://example.com/api/v2/long/path/resource.json"
result = abbreviate_path(url, max_len=40)
assert "resource.json" in result
assert "example.com" in result
def test_short_url_unchanged(self):
url = "https://example.com/api"
assert abbreviate_path(url) == url