An insert in apply_changes can now say that it joined your text to an existing line. The tool has
always done a character splice: text lands at the end of the matched anchor, not at the end of
that anchor's line. Anchor beta, text X, result betaX. Text with no trailing newline runs into the
following line when the anchor ends its line: beta\n plus X becomes beta\nXgamma. Inline and
mid-token insertion are legitimate uses, so the semantics stay; what is new is the warning, naming the
file, the line and the missing newline, on a dry run and on a write alike, and schema descriptions for
anchor and text naming where text lands. Silence was the defect: in .env, YAML, CSV, a Makefile
or a commented-out setting both outcomes are valid text, so you learn late. It surfaced reading a
preview to the end of each + line during nftables work.
The audit continues 1.14.0's description-budget work: every claim in every tool description, checked
against the code, one at a time. Three claims were false. run_python said only
printed output comes back (stderr is appended; the behaviour was right, the text was not). glob
claims to skip vendor directories it does not skip. search_content's max_results reads as a cap on
matches, but it caps contiguous match blocks: one run of adjacent hits is one block however long, so
max_results=5 over a dense log returns one long block.
Two behaviours were wrong. read_file counted a trailing newline as an extra line: a file ending in
\n reported N+1 lines, and offset=N+1 gave a phantom empty line, not "past the end". Counts now
match wc -l, and whole-file reads stay byte-identical. A capped single-file search_content dropped
its truncation flag, so five blocks came back with no sign the rest of the log existed. It ends with
(results truncated at N match blocks; raise max_results or narrow the pattern). The sudo rule,
dry_run's diff behaviour and expected_matches were each stated twice, so the wording was tightened:
58 fewer tokens on every request.
The sudo privilege check is a small shell lexical pass with two holes. A here-string (<<<)
opened no body, and <<-EOF named its delimiter -EOF, so in either case everything after it read as
heredoc text. A real sudo after cat <<< "$x" was never seen, so the command ran with no askpass
path and failed at the password prompt with nothing explaining why. The opposite also happened: a
heredoc quoting that command made a plain message prompt. All three editions now agree case for case,
pinned by one shared 17-case table (test_sudo_span_parser_matches_the_shared_case_table,
sudo_span_parser_matches_the_python_case_table, sudoGuardUnderstandsHeredocsAndHereStrings), the
rule written into spec.md under "sudo support". Suite green: 1157 passed, 8 skipped.