MeowOS-For-R3P2/patch_v2/reader_app/test/chapter_smoke.c

/* test/chapter_smoke.c — os.meow.reader 章节规范解析器宿主单测（纯 C，无 SDK/Freetype）。
 *
 * 覆盖（按需求 5 样本 + 跨段 + 中文数字转换）：
 *   C1 标准中文数字章（第一章/第二章/第十一章 → 1/2/11，char_off 单调）
 *   C2 阿拉伯章（第1章/第2章/第3章 → 1/2/3）
 *   C3 楔子/序章/尾声/番外（无号，number==-1，char_off 单调）
 *   C4 无章节纯文本（n==0，seg_char_start 仍完整）
 *   C5 开头长书名+简介再进第一章（卷首识别 as entries[0]、章号、防误判正文行 ）
 *   C6 跨段：>512KB 书，章落在后段（seg_idx>=1，char_off 跨段单调）
 *   C7 中文数字转换：十一/二十三/一百零一/一千零一/两
 *
 * 跑法：build_reader.sh 内 `test/chapter_smoke_host`；本文件只依赖 rdchapter + seg + rdenc。
 */
#include "rdchapter.h"
#include "seg.h"
#include "rdenc.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

static int g_pass = 0, g_fail = 0;
#define CHECK(cond, msg) do { if (cond) { g_pass++; } else { g_fail++; printf("FAIL %s:%d  %s\n", __FILE__, __LINE__, msg); } } while (0)

static char g_dir[600];

static void write_file(const char *path, const char *data)
{
    FILE *f = fopen(path, "wb");
    fwrite(data, 1, strlen(data), f);
    fclose(f);
}

/* 建段目录 + 切段，返回段数（<0=错误）。 */
static int seg_for(const char *path, const char *base)
{
    rdseg_set_base(base);
    char dir[600];
    if (rdseg_dir_for(path, dir, sizeof dir) != 0) return -1;
    strcpy(g_dir, dir);
    return rdseg_build(path, RD_ENC_UTF8, dir);
}

static int scan_book(rdchapter *c, const char *basename)
{
    return rdchapter_scan(g_dir, RD_ENC_UTF8, basename, c);
}

static void test_c1(void)
{
    const char *txt = "/tmp/ch_s1.txt";
    write_file(txt,
        "第一章 风起\n"
        "这是第一章的正文内容。\n"
        "第二章 云涌\n"
        "这是第二章的正文内容。\n"
        "第十一章 破晓\n"
        "这是第十一章的正文内容。\n");
    CHECK(seg_for(txt, "/tmp/ch_base1") >= 1, "c1 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s1") == 0, "c1 scan");
    CHECK(c.n == 3, "c1 n==3");
    if (c.n == 3) {
        CHECK(c.entries[0].number == 1, "c1 #1");
        CHECK(c.entries[1].number == 2, "c1 #2");
        CHECK(c.entries[2].number == 11, "c1 #11");
        CHECK(c.entries[0].char_off == 0, "c1 first at 0 (no preface)");
        CHECK(c.entries[0].char_off < c.entries[1].char_off &&
              c.entries[1].char_off < c.entries[2].char_off, "c1 char_off monotonic");
        CHECK(c.entries[0].seg_idx == 0 && c.entries[0].seg_char == 0, "c1 seg0 char0");
        CHECK(strcmp(c.entries[0].title, "第一章 风起") == 0, "c1 title");
        CHECK(c.has_pre == 0, "c1 no preface");
    }
    rdchapter_free(&c);
    remove(txt);
}

static void test_c2(void)
{
    const char *txt = "/tmp/ch_s2.txt";
    write_file(txt,
        "第1章 起源\n"
        "正文A。\n"
        "第2章 发展\n"
        "正文B。\n"
        "第3章 终局\n"
        "正文C。\n");
    CHECK(seg_for(txt, "/tmp/ch_base2") >= 1, "c2 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s2") == 0, "c2 scan");
    CHECK(c.n == 3, "c2 n==3");
    if (c.n == 3) {
        CHECK(c.entries[0].number == 1, "c2 #1");
        CHECK(c.entries[1].number == 2, "c2 #2");
        CHECK(c.entries[2].number == 3, "c2 #3");
        CHECK(c.entries[0].char_off < c.entries[1].char_off &&
              c.entries[1].char_off < c.entries[2].char_off, "c2 char_off monotonic");
    }
    rdchapter_free(&c);
    remove(txt);
}

static void test_c3(void)
{
    const char *txt = "/tmp/ch_s3.txt";
    write_file(txt,
        "楔子\n"
        "一段开篇。\n"
        "第一章 正传\n"
        "正文内容。\n"
        "尾声\n"
        "后记文字。\n"
        "番外 归途\n"
        "番外正文。\n");
    CHECK(seg_for(txt, "/tmp/ch_base3") >= 1, "c3 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s3") == 0, "c3 scan");
    CHECK(c.n == 4, "c3 n==4");
    if (c.n == 4) {
        CHECK(c.entries[0].number == -1, "c3 楔子 number -1");
        CHECK(strcmp(c.entries[0].title, "楔子") == 0, "c3 楔子 title");
        CHECK(c.entries[1].number == 1, "c3 第一章 number 1");
        CHECK(c.entries[2].number == -1, "c3 尾声 number -1");
        CHECK(c.entries[3].number == -1, "c3 番外 number -1");
        CHECK(c.entries[0].char_off < c.entries[1].char_off &&
              c.entries[1].char_off < c.entries[2].char_off &&
              c.entries[2].char_off < c.entries[3].char_off, "c3 char_off monotonic");
    }
    rdchapter_free(&c);
    remove(txt);
}

static void test_c4(void)
{
    const char *txt = "/tmp/ch_s4.txt";
    write_file(txt,
        "这是一个没有章节标题的长篇小说文本。\n"
        "它只有连续不断的段落，没有任何章回节的标识。\n"
        "所有的文字都是纯粹的故事内容。\n");
    CHECK(seg_for(txt, "/tmp/ch_base4") >= 1, "c4 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s4") == 0, "c4 scan");
    CHECK(c.n == 0, "c4 no chapters");
    CHECK(c.nsegs == 1, "c4 nsegs==1");
    CHECK(c.seg_char_start && c.seg_char_start[0] == 0 && c.seg_char_start[1] > 0, "c4 seg_char_start");
    rdchapter_free(&c);
    remove(txt);
}

static void test_c5(void)
{
    const char *txt = "/tmp/ch_s5.txt";
    write_file(txt,
        "《大漠孤烟》这是一本很长的书名，作者著。\n"
        "简介：这是一个故事的开头，讲述少年成长。\n"
        "第一章风雨欲来，全书开场，命运由此转折。\n"
        "第一章 离家\n"
        "正文内容。\n"
        "第二章 远行\n"
        "正文内容。\n");
    CHECK(seg_for(txt, "/tmp/ch_base5") >= 1, "c5 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "sample5") == 0, "c5 scan");
    CHECK(c.n == 3, "c5 n==3 (preface + 2 chapters)");
    if (c.n == 3) {
        CHECK(c.has_pre == 1, "c5 has preface");
        CHECK(c.entries[0].number == -1, "c5 preface number -1");
        CHECK(strcmp(c.entries[0].title, "sample5") == 0, "c5 preface title = basename");
        CHECK(c.entries[0].char_off == 0, "c5 preface char_off 0");
        CHECK(c.entries[1].number == 1 && c.entries[2].number == 2, "c5 chapter numbers 1/2");
        CHECK(c.entries[1].char_off > 0, "c5 ch1 after preface");
        CHECK(c.entries[0].char_off < c.entries[1].char_off &&
              c.entries[1].char_off < c.entries[2].char_off, "c5 monotonic");
        /* 防误判：正文行「第一章风雨欲来，全书开场。」带两个正文标点 → 未被识别为章 */
        CHECK(c.entries[1].seg_char > 0, "c5 decoy line not counted (ch1 seg_char>0)");
    }
    rdchapter_free(&c);
    remove(txt);
}

static void test_c6(void)
{
    /* 跨段：写 ~600KB filler，在 ~600KB 处放一个章 → 应落在后段（seg_idx>=1） */
    const char *txt = "/tmp/ch_s6.txt";
    FILE *f = fopen(txt, "wb");
    const char *para = "这是一段用于填充页面的中文正文，包含标点，中文English混合，反复出现，占据空间。\n";
    size_t plen = strlen(para);
    size_t total = 0;
    while (total < 600 * 1024) {
        fwrite(para, 1, plen, f); total += plen;
    }
    fwrite("第二百章 跨段检测\n", 1, strlen("第二百章 跨段检测\n"), f);
    const char *tail = "这是跨段之后的一小段正文。\n";
    fwrite(tail, 1, strlen(tail), f);
    fclose(f);
    CHECK(seg_for(txt, "/tmp/ch_base6") >= 2, "c6 seg build >=2");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s6") == 0, "c6 scan");
    /* 前置 filler（600KB 无章节）→ entries[0]=卷首；真正章节在后段 */
    CHECK(c.n == 2, "c6 n==2 (preface + 1 chapter)");
    if (c.n == 2) {
        CHECK(c.entries[0].number == -1, "c6 preface number -1");
        CHECK(c.entries[1].number == 200, "c6 number 200");
        CHECK(c.entries[1].seg_idx >= 1, "c6 chapter in later segment");
        CHECK(c.entries[1].seg_idx < c.nsegs, "c6 seg_idx in range");
        CHECK(c.entries[1].seg_char >= 0, "c6 seg_char >=0");
        CHECK(c.seg_char_start[0] == 0 &&
              c.seg_char_start[0] < c.seg_char_start[1] &&
              c.seg_char_start[c.nsegs - 1] < c.seg_char_start[c.nsegs], "c6 seg_char_start monotonic");
        /* 跨段 char_off：章首偏移应在段 1 起点之后 */
        CHECK(c.entries[1].char_off >= c.seg_char_start[c.entries[1].seg_idx], "c6 char_off within seg");
    }
    rdchapter_free(&c);
    remove(txt);
}

static void test_c7(void)
{
    CHECK(rdchapter_cn_num("十一", 6) == 11, "c7 十一=11");
    CHECK(rdchapter_cn_num("二十三", 9) == 23, "c7 二十三=23");
    CHECK(rdchapter_cn_num("一百零一", 12) == 101, "c7 一百零一=101");
    CHECK(rdchapter_cn_num("一千零一", 12) == 1001, "c7 一千零一=1001");
    CHECK(rdchapter_cn_num("两", 3) == 2, "c7 两=2");
    CHECK(rdchapter_cn_num("一百", 6) == 100, "c7 一百=100");
    CHECK(rdchapter_cn_num("foobar", 6) == -1, "c7 invalid -1");
}

static void test_c8(void)
{
    /* R3.4 英文章节：文字数字（one/twenty-one/two hundred fifty）+ 罗马（IV）+ 拒绝无空格（ChapterX） */
    const char *txt = "/tmp/ch_s8.txt";
    write_file(txt,
        "Chapter one\n"
        "正文。\n"
        "Chapter IV\n"
        "正文。\n"
        "Chapter twenty-one\n"
        "正文。\n"
        "Chapter two hundred fifty\n"
        "正文。\n"
        "ChapterX\n"
        "正文。\n"
        "Chapter twelve\n"
        "正文。\n");
    CHECK(seg_for(txt, "/tmp/ch_base8") >= 1, "c8 seg build");
    rdchapter c; rdchapter_init(&c);
    CHECK(scan_book(&c, "s8") == 0, "c8 scan");
    /* ChapterX 不认（无空格）；其余 5 个认 */
    CHECK(c.n == 5, "c8 n==5");
    if (c.n == 5) {
        static const int exp[5] = { 1, 4, 21, 250, 12 };
        for (int i = 0; i < 5; i++) CHECK(c.entries[i].number == exp[i], "c8 word number value");
        CHECK(c.entries[0].char_off < c.entries[1].char_off &&
              c.entries[1].char_off < c.entries[2].char_off &&
              c.entries[2].char_off < c.entries[3].char_off &&
              c.entries[3].char_off < c.entries[4].char_off, "c8 char_off monotonic");
        CHECK(strcmp(c.entries[0].title, "Chapter one") == 0, "c8 title Chapter one");
        CHECK(strcmp(c.entries[3].title, "Chapter two hundred fifty") == 0, "c8 title two hundred fifty");
    }
    rdchapter_free(&c);
    remove(txt);
}

int main(void)
{
    test_c1();
    test_c2();
    test_c3();
    test_c4();
    test_c5();
    test_c6();
    test_c7();
    test_c8();
    printf("chapter_smoke: %d pass, %d fail\n", g_pass, g_fail);
    return g_fail ? 1 : 0;
}