feat(xce-test-cli-ux): add flag typo-suggestion and help-block renderer

flagspec_reject() now names a likely-intended flag when an unrecognized
flag closely matches exactly one entry in the descriptor table (genuine
Levenshtein DP, silent on zero or ambiguous matches). flagspec_render_help()
generates the short/long/required-optional/default flag-listing block from
a descriptor table. 1278/1278 controller suite passing (+5).

Task 1.2 of .kiro/specs/xce-test-cli-ux/tasks.md.
Requirements: 6.1, 6.2, 6.3, 6.4, 8.2, 10.1, 10.2, 10.3

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019jaMf9QEwpugH2W9KAuc7f
This commit is contained in:
Rongsong Shen 2026-09-09 07:40:46 +00:00
parent 0748f10cc6
commit 89af46af3e
No known key found for this signature in database
GPG key ID: 5BCDAAE0123E9FE2
4 changed files with 335 additions and 17 deletions

View file

@ -9,7 +9,7 @@
- _Requirements: 2.1, 2.2, 2.3, 2.4, 3.1_
- _Boundary: Flag Parsing Layer_
- [ ] 1.2 Build the flag typo-suggestion utility and the generated help-flag-block renderer
- [x] 1.2 Build the flag typo-suggestion utility and the generated help-flag-block renderer
- Add a small edit-distance comparison so an unrecognized flag that closely matches exactly one flag in the descriptor table is named as the likely intended one in the rejection message
- Add a renderer that produces the flag-listing portion of help text (short form, long form, required/optional, default) directly from the descriptor table
- Observable completion: the unit test suite shows a single close-match flag typo producing a "did you mean" suggestion, a non-close-match producing no suggestion, and the renderer producing a correctly formatted block for a multi-flag test table

View file

@ -3,6 +3,7 @@
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
/* True iff `token` (a raw command-line argument, e.g. "-h", "--host", or
@ -80,6 +81,139 @@ int flagspec_build_opts(const flagspec_t* spec, size_t n,
return 0;
}
/* Levenshtein edit distance between two NUL-terminated strings -- the
* minimum number of single-character insertions/deletions/substitutions
* that turns `a` into `b`. Small and self-contained: research.md confirms
* no edit-distance utility exists anywhere in the codebase yet. Standard
* two-row rolling DP, O(len(a)*len(b)) time, O(min(len(a),len(b))+1)
* space -- more than fast enough for flag-name-sized strings.
*
* Returns (size_t)-1 on allocation failure, which callers must treat as
* "not a close match" rather than a false zero-distance/exact-match
* signal. */
static size_t levenshtein(const char* a, const char* b)
{
size_t la = strlen(a), lb = strlen(b);
/* Keep the rolling row sized to the SHORTER string. */
if (lb > la) {
const char* tmp = a;
a = b;
b = tmp;
size_t t = la;
la = lb;
lb = t;
}
size_t* row = calloc(lb + 1, sizeof(size_t));
if (!row)
return (size_t)-1;
for (size_t j = 0; j <= lb; j++)
row[j] = j;
for (size_t i = 1; i <= la; i++) {
size_t prev_diag = row[0];
row[0] = i;
for (size_t j = 1; j <= lb; j++) {
size_t saved = row[j];
size_t cost = (a[i - 1] == b[j - 1]) ? 0 : 1;
size_t del = row[j] + 1;
size_t ins = row[j - 1] + 1;
size_t sub = prev_diag + cost;
size_t best = del < ins ? del : ins;
if (sub < best)
best = sub;
row[j] = best;
prev_diag = saved;
}
}
size_t result = row[lb];
free(row);
return result;
}
/* True iff `candidate` (one flag's literal spelling, e.g. "-h" or
* "--host") "closely matches" `bad_flag` for typo-suggestion purposes
* (Requirement 10.1).
*
* Threshold, a deliberate design choice (design.md/research.md do not
* mandate a specific algorithm or cutoff): floor(len(candidate)/4),
* clamped to [1,3]. A fixed absolute threshold would be wrong at both
* ends -- distance 1 from a 2-character short flag ("-h" vs "-z") is a
* near-total rewrite, while distance 1 from an 8+ character long flag is
* a single typo -- so the tolerance scales with the spelling's own
* length instead. Exact matches (distance 0) are never "close" here:
* flagspec_reject() only calls this from the already-unrecognized branch,
* where an exact-spelling candidate would be a contradiction, and this
* keeps the helper correct standalone too. */
static int is_close_match(const char* bad_flag, const char* candidate)
{
size_t dist = levenshtein(bad_flag, candidate);
if (dist == 0 || dist == (size_t)-1)
return 0;
size_t len = strlen(candidate);
size_t threshold = len / 4;
if (threshold < 1)
threshold = 1;
if (threshold > 3)
threshold = 3;
return dist <= threshold;
}
/* Writes `fs`'s display label ("-h/--host" or, for a long-only flag,
* "--json") into `buf`. */
static void format_flag_label(char* buf, size_t cap, const flagspec_t* fs)
{
if (fs->short_opt)
snprintf(buf, cap, "-%c/--%s", fs->short_opt, fs->long_opt);
else
snprintf(buf, cap, "--%s", fs->long_opt);
}
/* Returns the single flagspec_t in `spec` whose short or long spelling
* closely matches `bad_flag`, or NULL when zero or more than one entry
* qualifies -- Requirement 10.1 only promises a guess when exactly one
* flag is a plausible intended target; an ambiguous typo (two or more
* equally-plausible candidates) gets no guess at all, the same as a typo
* that matches nothing (Requirement 10.2). */
static const flagspec_t* find_typo_suggestion(const flagspec_t* spec, size_t n,
const char* bad_flag)
{
if (!spec || !bad_flag || !*bad_flag)
return NULL;
const flagspec_t* found = NULL;
for (size_t i = 0; i < n; i++) {
const flagspec_t* fs = &spec[i];
int is_close = 0;
if (fs->short_opt) {
char short_form[3] = { '-', fs->short_opt, '\0' };
if (is_close_match(bad_flag, short_form))
is_close = 1;
}
if (!is_close && fs->long_opt && *fs->long_opt) {
char long_form[128];
snprintf(long_form, sizeof(long_form), "--%s", fs->long_opt);
if (is_close_match(bad_flag, long_form))
is_close = 1;
}
if (is_close) {
if (found)
return NULL; /* ambiguous -- more than one candidate */
found = fs;
}
}
return found;
}
int flagspec_reject(const char* cmd_name, const flagspec_t* spec, size_t n,
const char* bad_flag, const char* usage_text)
{
@ -103,8 +237,18 @@ int flagspec_reject(const char* cmd_name, const flagspec_t* spec, size_t n,
fprintf(stderr, "xce_test: %s: flag '%s' does not take a value\n",
who, display);
} else {
fprintf(stderr, "xce_test: %s: unrecognized flag '%s'\n", who,
display);
const flagspec_t* suggestion = find_typo_suggestion(spec, n, bad_flag);
if (suggestion) {
char label[144];
format_flag_label(label, sizeof(label), suggestion);
fprintf(stderr,
"xce_test: %s: unrecognized flag '%s' (did you mean "
"'%s'?)\n",
who, display, label);
} else {
fprintf(stderr, "xce_test: %s: unrecognized flag '%s'\n", who,
display);
}
}
if (usage_text && *usage_text) {
@ -117,3 +261,38 @@ int flagspec_reject(const char* cmd_name, const flagspec_t* spec, size_t n,
return -EINVAL;
}
char* flagspec_render_help(const flagspec_t* spec, size_t n)
{
if (!spec || n == 0)
return NULL;
char* buf = NULL;
size_t size = 0;
FILE* f = open_memstream(&buf, &size);
if (!f)
return NULL;
for (size_t i = 0; i < n; i++) {
const flagspec_t* fs = &spec[i];
if (fs->short_opt)
fprintf(f, " -%c, --%s", fs->short_opt, fs->long_opt);
else
fprintf(f, " --%s", fs->long_opt);
fprintf(f, " %s", fs->required ? "required" : "optional");
if (fs->default_desc && *fs->default_desc)
fprintf(f, " (default: %s)", fs->default_desc);
fprintf(f, "\n");
}
if (fclose(f) != 0) {
free(buf);
return NULL;
}
return buf;
}

View file

@ -1,24 +1,26 @@
#pragma once
/*
* flagspec.h -- flag-descriptor table format and getopt_long-based parsing
* core for xce_test's command-line surface (xce-test-cli-ux spec, task 1.1;
* Requirements 2.1-2.4, 3.1; the flagspec_t shape is a stable contract also
* relied on by task 1.2 (typo suggestion, flagspec_render_help) and task
* 3.1 (migrating every existing cmd_* command onto this table)).
* flagspec.h -- flag-descriptor table format, getopt_long-based parsing
* core, typo suggestion, and help-flag-block rendering for xce_test's
* command-line surface (xce-test-cli-ux spec, tasks 1.1/1.2; Requirements
* 2.1-2.4, 3.1, 6.1-6.4, 8.2, 10.1-10.3; the flagspec_t shape is a stable
* contract also relied on by task 3.1 (migrating every existing cmd_*
* command onto this table)).
*
* bug-1711's root cause was that xce_test's flag parsing had no shared
* helper at all: a plain getopt() loop per command, each with its own
* hand-written switch/default, drifted independently from its
* hand-authored USAGE_* string. This module gives every flag-accepting
* command ONE descriptor table to build its getopt_long() option table
* from and ONE shared rejection path to call on any parse failure, so the
* table is the single source of truth for both.
* from, ONE shared rejection path (with typo suggestion) to call on any
* parse failure, and ONE renderer for the flag-listing portion of its
* help text, so the table is the single source of truth for all three.
*
* THIS FILE/TASK (1.1) declares and implements only flagspec_build_opts()
* and flagspec_reject(). flagspec_render_help() (the help-text flag-block
* renderer) and flagspec_reject()'s typo-suggestion behavior are task
* 1.2 and are intentionally NOT declared here yet -- task 1.2 extends
* this header when it adds them.
* Task 1.1 declared/implemented flagspec_build_opts() and
* flagspec_reject() (without typo suggestion). THIS TASK (1.2) adds
* flagspec_render_help() and extends flagspec_reject()'s rejection
* message with a "did you mean" suggestion (Requirement 10) when exactly
* one flag in `spec` closely matches the unrecognized token.
*/
#include <getopt.h>
@ -116,8 +118,16 @@ int flagspec_build_opts(const flagspec_t* spec, size_t n,
* 2.4) to stderr only (Requirement 3.1 -- never stdout) and returns a
* negative value; it never partially executes the command.
*
* Typo suggestion ("did you mean --host?", Requirement 10) is task 1.2,
* not implemented by this function yet.
* Typo suggestion (Requirement 10): when `bad_flag` matches no entry in
* `spec` (the "unrecognized flag" case above) and exactly one entry's
* short or long spelling closely matches it by edit distance (see
* flagspec.c's is_close_match()), the rejection message names that flag
* as the likely intended one (Requirement 10.1). When zero or more than
* one entry qualifies as a close match, no guess is made (Requirement
* 10.2) -- an ambiguous typo is treated the same as a typo that matches
* nothing at all. Suggestion is never attempted for the "requires a
* value"/"does not take a value" cases, since those already identify the
* exact intended flag.
*
* `usage_text` is expected to be the command's one-line usage summary
* (e.g. "jobkill -j <jobid> [-s <signo>]"), not its full help block; it
@ -128,3 +138,17 @@ int flagspec_build_opts(const flagspec_t* spec, size_t n,
* case. */
int flagspec_reject(const char* cmd_name, const flagspec_t* spec, size_t n,
const char* bad_flag, const char* usage_text);
/* Renders the flag-listing portion of a command's help text (Requirement
* 8.2) directly from `spec`: one line per flag, each stating its short
* form (when it has one), its long form, whether it is required or
* optional, and its default-value description (when `default_desc` is
* non-NULL). This is the flag-block ONLY -- the surrounding description,
* example, and related-command prose (Requirement 8.1/8.3/8.4) stay
* hand-authored, per design.md's "Flag-descriptor table as the single
* source of truth" decision.
*
* Returns a heap-allocated, NUL-terminated string the caller must free().
* Returns NULL if `spec` is NULL or n == 0, or if the underlying
* allocation fails; it never returns a partially-written buffer. */
char* flagspec_render_help(const flagspec_t* spec, size_t n);

View file

@ -375,6 +375,116 @@ static void test_reject_no_argument_flag_given_a_value_reports_distinctly(void)
cleanup_dir();
}
/* ---- flagspec_reject() typo-suggestion tests (Requirement 10) ---------
* A second fixture, g_ambiguous_spec, is purpose-built for the
* ambiguous-match case below so its edit distances are deliberate rather
* than incidental to g_testcmd_spec's two very dissimilar flag names. */
static const flagspec_t g_ambiguous_spec[2] = {
{ 0, "targeta", "target-a", no_argument, 0, NULL },
{ 0, "targetb", "target-b", no_argument, 0, NULL },
};
static void test_reject_suggests_single_close_match_flag_typo(void)
{
/* "--hst" is "--host" with the 'o' deleted -- one edit, well inside
* the closeness threshold for a 6-character long-form spelling -- and
* is nowhere near "--command". Requirement 10.1: name the
* likely-intended flag in the rejection message. */
fresh_dir();
capture_t cap;
capture_begin(&cap);
int rc = flagspec_reject("testcmd", g_testcmd_spec, 2, "--hst",
g_testcmd_usage);
char out_buf[512], err_buf[512];
capture_end(&cap, out_buf, sizeof(out_buf), err_buf, sizeof(err_buf));
TEST_ASSERT_TRUE(rc < 0);
TEST_ASSERT_EQUAL_STRING("", out_buf);
TEST_ASSERT_NOT_NULL(strstr(err_buf, "unrecognized flag '--hst'"));
TEST_ASSERT_NOT_NULL(strstr(err_buf, "did you mean"));
TEST_ASSERT_NOT_NULL(strstr(err_buf, "--host"));
cleanup_dir();
}
static void test_reject_gives_no_suggestion_for_a_distant_typo(void)
{
/* "--wombat" does not closely resemble "--host" or "--command" by any
* reasonable edit-distance threshold. Requirement 10.2: report
* unrecognized without guessing. */
fresh_dir();
capture_t cap;
capture_begin(&cap);
int rc = flagspec_reject("testcmd", g_testcmd_spec, 2, "--wombat",
g_testcmd_usage);
char out_buf[512], err_buf[512];
capture_end(&cap, out_buf, sizeof(out_buf), err_buf, sizeof(err_buf));
TEST_ASSERT_TRUE(rc < 0);
TEST_ASSERT_EQUAL_STRING("", out_buf);
TEST_ASSERT_NOT_NULL(strstr(err_buf, "unrecognized flag '--wombat'"));
TEST_ASSERT_NULL(strstr(err_buf, "did you mean"));
cleanup_dir();
}
static void test_reject_gives_no_suggestion_when_typo_is_ambiguous(void)
{
/* "--targetc" is exactly one substitution away from BOTH "--targeta"
* and "--targetb". Requirement 10.1 only promises a guess when
* exactly one flag qualifies -- two equally-plausible candidates must
* produce no suggestion at all, the same as a typo that matches
* nothing. */
fresh_dir();
capture_t cap;
capture_begin(&cap);
int rc = flagspec_reject("testcmd", g_ambiguous_spec, 2, "--targetc",
"testcmd --targeta|--targetb");
char out_buf[512], err_buf[512];
capture_end(&cap, out_buf, sizeof(out_buf), err_buf, sizeof(err_buf));
TEST_ASSERT_TRUE(rc < 0);
TEST_ASSERT_EQUAL_STRING("", out_buf);
TEST_ASSERT_NOT_NULL(strstr(err_buf, "unrecognized flag '--targetc'"));
TEST_ASSERT_NULL(strstr(err_buf, "did you mean"));
cleanup_dir();
}
/* ---- flagspec_render_help() tests (Requirement 8.2) -------------------- */
static void test_render_help_formats_multi_flag_table(void)
{
flagspec_t spec[3];
memcpy(spec, g_testcmd_spec, sizeof(g_testcmd_spec));
spec[2] = (flagspec_t){ 0, "json", "output-format", no_argument, 0,
"table" };
char* help = flagspec_render_help(spec, 3);
TEST_ASSERT_NOT_NULL(help);
/* required flag: short + long form, marked required */
TEST_ASSERT_NOT_NULL(strstr(help, "-h"));
TEST_ASSERT_NOT_NULL(strstr(help, "--host"));
TEST_ASSERT_NOT_NULL(strstr(help, "required"));
/* optional flag: short + long form, marked optional, default shown */
TEST_ASSERT_NOT_NULL(strstr(help, "-c"));
TEST_ASSERT_NOT_NULL(strstr(help, "--command"));
TEST_ASSERT_NOT_NULL(strstr(help, "optional"));
TEST_ASSERT_NOT_NULL(strstr(help, "off"));
/* long-only flag: no short form, still shows long form + default */
TEST_ASSERT_NOT_NULL(strstr(help, "--json"));
TEST_ASSERT_NOT_NULL(strstr(help, "table"));
free(help);
}
static void test_render_help_rejects_null_and_empty_inputs(void)
{
TEST_ASSERT_NULL(flagspec_render_help(NULL, 2));
TEST_ASSERT_NULL(flagspec_render_help(g_testcmd_spec, 0));
}
void test_flagspec(void)
{
UnitySetTestFile(__FILE__);
@ -390,4 +500,9 @@ void test_flagspec(void)
RUN_TEST(test_reject_rejects_missing_value_short_flag);
RUN_TEST(test_reject_handles_null_bad_flag_and_null_usage_gracefully);
RUN_TEST(test_reject_no_argument_flag_given_a_value_reports_distinctly);
RUN_TEST(test_reject_suggests_single_close_match_flag_typo);
RUN_TEST(test_reject_gives_no_suggestion_for_a_distant_typo);
RUN_TEST(test_reject_gives_no_suggestion_when_typo_is_ambiguous);
RUN_TEST(test_render_help_formats_multi_flag_table);
RUN_TEST(test_render_help_rejects_null_and_empty_inputs);
}