Documentation
¶
Overview ¶
Package textutil provides small, dependency-free string-safety primitives shared by packages that must not depend on each other: rune-safe byte-cap truncation (jschema, delivery) and control-byte detection (workflows/controller, workflows/compiler). It also provides ASCII-folded containment (ContainsFold). Each importer needs the exact same check; this leaf package holds the one implementation.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func ContainsFold ¶
ContainsFold reports whether s contains substr when only ASCII letters fold: 'A'-'Z' matches 'a'-'z' and 'a'-'z' matches 'A'-'Z'. Every other byte, including non-ASCII runes and malformed UTF-8, compares exactly and never folds. An empty substr is contained in every string. The scan uses a byte window of length len(substr) over s and is O(len(s)*len(substr)) in the worst case.
func HasControlByte ¶
HasControlByte reports whether s contains a C0 control byte (0x00-0x1F) or DEL (0x7F). Callers that build an identity or digest by concatenating untrusted strings with a byte-level separator use this to reject a value that could smuggle that separator and collide two different identities.
func TruncateEllipsis ¶
TruncateEllipsis truncates s to a byte budget and appends the ellipsis rune "…" (U+2026, 3 bytes) when truncation happens and the budget can fit the marker.
When len(s) <= maxBytes, it returns s unchanged with no marker. When maxBytes <= 0, it returns "". When maxBytes < 3, the marker cannot fit, so it falls back to TruncateRuneSafe(s, maxBytes) with no marker. Otherwise it appends "…" to the longest prefix of s that fits maxBytes-3 bytes on a UTF-8 rune boundary. The result never exceeds maxBytes. For valid UTF-8 input, the result is valid UTF-8.
func TruncateMiddle ¶
TruncateMiddle shortens s to at most maxLen runes by keeping a rune-safe prefix and suffix and inserting one ASCII "..." marker between them. maxLen counts runes via utf8.RuneCountInString, not bytes.
When the rune count of s is at most maxLen, it returns s unchanged with no marker. When maxLen <= 0, it returns "". When maxLen < 3, the marker cannot fit, so it returns the first maxLen runes with no marker. Otherwise it splits the remaining maxLen-3 runes evenly: the prefix gets ceil((maxLen-3)/2) runes and the suffix gets the rest.
The result never exceeds maxLen runes and never splits a rune. For valid UTF-8 input, the result is valid UTF-8.
func TruncateRuneSafe ¶
TruncateRuneSafe returns the longest prefix of s that is at most maxBytes bytes and ends on a UTF-8 rune boundary.
When len(s) <= maxBytes, it returns s unchanged. When maxBytes <= 0, it returns "". A cut inside a rune backs off to the previous rune start, so the result is always valid UTF-8 for valid UTF-8 input.
func TruncateTail ¶
TruncateTail returns the longest suffix of s that is at most maxBytes bytes and starts on a UTF-8 rune boundary.
When len(s) <= maxBytes, it returns s unchanged. When maxBytes <= 0, it returns "". Error chains read wrapper first and root cause last, so the tail preserves the root cause for diagnostics.
Types ¶
This section is empty.
Source Files