feat: implement new streecmp loose checking

.gitignore

@@ -1,3 +1,3 @@
 *.o
 compile_commands.json
-tinyparser
+test

Makefile

@@ -1,14 +1,26 @@
 CC=gcc
+LD=ld
 CFLAGS=-g3
 CFILES=parslib.c strings.c
-CFILES_STREECMP=streecmp/streecmp.c
-OUTPUT=parslib
+TEST_FILES=test.c
+OUTPUT=parslib.o
+TOUTPUT=test
+LOUTPUT=parslib.final.o
 
-all: 
-	$(CC) $(CFLAGS) $(CFILES) $(CFILES_STREECMP) -o $(OUTPUT)
+all: $(FILES)
+	$(CC) -c $(CFLAGS) -o parslib.o parslib.c
+	$(CC) -c $(CFLAGS) -o strings.o strings.c
+	$(CC) -c $(CFLAGS) -o streecmp/streecmp.o streecmp/streecmp.c
+	$(LD) -relocatable -o $(LOUTPUT) parslib.o strings.o streecmp/streecmp.o
+
+test: all
+	$(CC) $(CFLAGS) $(TEST_FILES) parslib.o strings.o streecmp/streecmp.o -o $(TOUTPUT)
+
+run_test: 
+	./$(TOUTPUT)
 
 run: 
 	./$(OUTPUT)
 
 clean: 
-	rm -rf $(OUTPUT)
+	rm -rf *.o $(TOUTPUT) $(OUTPUT) $(LOUTPUT)

README

@@ -27,14 +27,10 @@ TECHNICALS
 
 LEAKS 
 
-    LEAKS file contains the most recent valgrind memory leaks 
-    dump ran on main.c 
-
-    LEAKS.TEST.[id] files respectively contain the most recent 
-    valgrind memory leaks dump ran on a test with an id=[id]
+    LEAKS file contains the most recent valgrind 
+    memory leaks dump ran on a test program.
 
 TODO 
     
-    * implement tests 
     * integrations with tinyproxy..?
 

parslib.c

@@ -10,19 +10,19 @@ extern char *headers;
 static struct nod *method_tree = NULL;
 static struct nod *header_tree = NULL; 
 
-static int stoin(char *str, int len, int *out) {
+int stoin(char *str, int len, int *out) {
     int ret = 0; 
     int place = 1; 
-    for (char *chr = str+len; chr >= str; chr--) {
+    for (char *chr = str+len-1; chr >= str; chr--) {
         if (chr == str) {
-            if (*ch == '+') {
+            if (*chr == '+') {
                 goto _proceed;
             }
-            if (*ch == '-') {
+            if (*chr == '-') {
                 goto _proceed;
             }
         }
-        if (*chr >= '0' && *ch <= '9') {
+        if (*chr >= '0' && *chr <= '9') {
             goto _proceed;
         }
 
@@ -44,21 +44,43 @@ _proceed:
     }
 
     *out = ret;
+    return 0;
 }
 
-void printfpars(struct httpars *pars) {
+void printfpareq(struct httpareq *req) {
     fprintf(stderr, "\tstats:\n"
                     "\t\tmethod\t: %d\n"
                     "\t\turi\t: %.*s\n"
                     "\t\tver\t: %.*s\n",
-                    pars->titl.method, 
-                    pars->titl.uri.len, pars->titl.uri.er, 
-                    pars->titl.ver.len, pars->titl.ver.er
+                    req->titl.method, 
+                    req->titl.uri.len, req->titl.uri.er, 
+                    req->titl.ver.len, req->titl.ver.er
             );
 
     fprintf(stdout, "\theaders:\n");
     for (int i = 0; i < header_count; i++) {
-        struct point *pnt = &pars->hentries[i]; 
+        struct point *pnt = &req->hentries[i]; 
+        if (!pnt->er) {
+            continue;
+        }
+
+        fprintf(stdout, "\t\t%d\t: %.*s\n", i, pnt->len, pnt->er);
+    }
+}
+
+void printfpares(struct httpares *res) {
+    fprintf(stderr, "\tstats:\n"
+                    "\t\tver\t: %.*s\n"
+                    "\t\tcode\t: %d\n"
+                    "\t\tstext\t: %.*s\n",
+                    res->titl.ver.len, res->titl.ver.er,
+                    res->titl.code, 
+                    res->titl.stxt.len, res->titl.stxt.er
+            );
+
+    fprintf(stdout, "\theaders:\n");
+    for (int i = 0; i < header_count; i++) {
+        struct point *pnt = &res->hentries[i]; 
         if (!pnt->er) {
             continue;
         }
@@ -80,12 +102,12 @@ int initres(void) {
         return -1;
     }
 
-    ret = gentree(header_tree, headers, NULL);
+    ret = gentree(header_tree, headers, NULL, LOOSE_CHECK);
     if (ret < 0) {
         return -1;
     }
 
-    ret = gentree(method_tree, methods, NULL);
+    ret = gentree(method_tree, methods, NULL, LOOSE_CHECK);
     if (ret < 0) {
         return -1;
     }
@@ -94,8 +116,44 @@ int initres(void) {
 }
 
 void fretres(void) {
-    free(method_tree);
-    free(header_tree);
+    frenod(method_tree);
+    frenod(header_tree);
+}
+
+void frepareq(struct httpareq *req) {
+    struct httitlereq *titl = &req->titl;
+    free(titl->uri.er);
+    titl->uri.er = NULL;
+    free(titl->ver.er);
+    titl->uri.er = NULL;
+
+    for (int i = 0; i < header_count; i++) {
+        struct point *e = &req->hentries[i];
+        if (!e->er) {
+            continue;
+        }
+
+        free(e->er);
+        e->er = NULL;
+    }
+}
+
+void frepares(struct httpares *res) {
+    struct httitleres *titl = &res->titl;
+    free(titl->ver.er);
+    titl->ver.er = NULL;
+    free(titl->stxt.er);
+    titl->stxt.er = NULL;
+
+    for (int i = 0; i < header_count; i++) {
+        struct point *e = &res->hentries[i];
+        if (!e->er) {
+            continue;
+        }
+
+        free(e->er);
+        e->er = NULL;
+    }
 }
 
 int readlin(char **buff, char **buff_lim) {
@@ -134,9 +192,9 @@ int parshfield(char *offset, int len, struct point *hentries) {
     }
 
     diff = htitle_lim-cursor;
-    ret = streencmp(header_tree, cursor, diff);
+    ret = streencmp(header_tree, cursor, diff, LOOSE_CHECK);
     if (!ret) {
-        return -1;
+        return 0; // skip it 
     }
 
     key = ret;
@@ -158,13 +216,139 @@ _loop:
 
     // header value
     diff = cursor_lim-cursor;
+    char *e = strndup(cursor, diff);
+    if (!e) {
+        return -1;
+    }
 
-    hentries[key].er = cursor;
+    hentries[key].er = e;
     hentries[key].len = diff;
 
     return 0;
 }
 
+int pahostinfo(char *offset, int len, struct hostinfo *info) {
+    int diff = 0; 
+    int usedefault = 0;
+    char *delim = strchr(offset, ':');
+    if (!delim) {
+        delim = offset+len;
+        usedefault = 1;
+    }
+
+    diff = delim-offset;
+    info->hostname = strndup(offset, diff);
+    if (!info->hostname) {
+        goto _err;
+    }
+    info->hostname_len = diff;
+
+    if (usedefault) {
+        diff = strlen(DEFAULT_HTTP);
+        info->service = strndup(DEFAULT_HTTP, diff);
+        if (!info->service) {
+            goto _err_service;
+        }
+        info->service_len = diff;
+    } else {
+        char *service_offset = delim+1;
+        char *end = offset+len;
+        diff = end-service_offset;
+        info->service = strndup(service_offset, diff);
+        if (!info->service) {
+            goto _err_service;
+        }
+        info->service_len = diff;
+    }
+
+    return 0;
+
+_err_service:
+    free(info->hostname);
+
+_err:
+    return -1;
+}
+
+int pareqtitl(char *offset, int len, struct httitlereq *titl) {
+    int ret = 0; 
+    int diff = 0; 
+    char *cursor = offset;
+    char *cursor_lim = cursor+len;
+
+    // method
+    char *method_lim = strchr(cursor, ' ');
+    if (!method_lim) {
+        return -1;
+    }
+    if (method_lim > cursor_lim) {
+        return -1;
+    }
+
+    diff = method_lim-cursor;
+    ret = streencmp(method_tree, cursor, diff, LOOSE_CHECK);
+    if (ret == 0) {
+        return -1;
+    }
+
+    titl->method = ret;
+    cursor += diff;
+
+    // white space 
+_loop1: 
+    if (cursor > cursor_lim) {
+        return -1;
+    }
+    if (*cursor == ' ') {
+        cursor++;
+        goto _loop1;
+    }
+
+    // uri 
+    char *uri_lim = strchr(cursor, ' ');
+    if (!uri_lim) {
+        return -1;
+    }
+    if (uri_lim > cursor_lim) {
+        return -1;
+    }
+
+    diff = uri_lim-cursor;
+    char *e = strndup(cursor, diff);
+    if (!e) {
+        return -1;
+    }
+
+    titl->uri.er = e;
+    titl->uri.len = diff;
+
+    cursor += diff;
+    
+    // white space 
+_loop2:
+    if (cursor > cursor_lim) {
+        return -1;
+    }
+    if (*cursor == ' ') {
+        cursor++;
+        goto _loop2;
+    }
+
+    // ver
+    diff = cursor_lim-cursor;
+    e = strndup(cursor, diff); 
+    if (!e) {
+        return -1;
+    }
+
+    titl->ver.er = e;
+    titl->ver.len = diff;
+
+    cursor += diff;
+
+    return 0;
+}
+
 int parestitl(char *offset, int len, struct httitleres *titl) {
     int ret = 0; 
     int diff = 0;
@@ -181,7 +365,12 @@ int parestitl(char *offset, int len, struct httitleres *titl) {
     }
 
     diff = ver_lim-cursor;
-    titl->ver.er = cursor;
+    char *e = strndup(cursor, diff); 
+    if (!e) {
+        return -1;
+    }
+
+    titl->ver.er = e;
     titl->ver.len = diff;
 
     cursor += diff;
@@ -227,7 +416,12 @@ _loop2:
 
     // status text
     diff = cursor_lim-cursor;
-    titl->stxt.er = cursor;
+    e = strndup(cursor, diff);
+    if (!e) {
+        return -1;
+    }
+
+    titl->stxt.er = e;
     titl->stxt.len = diff;
 
     cursor += diff;
@@ -235,82 +429,13 @@ _loop2:
     return 0;
 }
 
-int pareqtitl(char *offset, int len, struct httitlereq *titl) {
-    int ret = 0; 
-    int diff = 0; 
-    char *cursor = offset;
-    char *cursor_lim = cursor+len;
-
-    // method
-    char *method_lim = strchr(cursor, ' ');
-    if (!method_lim) {
-        return -1;
-    }
-    if (method_lim > cursor_lim) {
-        return -1;
-    }
-
-    diff = method_lim-cursor;
-    ret = streencmp(method_tree, cursor, diff);
-    if (ret == 0) {
-        return -1;
-    }
-
-    titl->method = ret;
-    cursor += diff;
-
-    // white space 
-_loop1: 
-    if (cursor > cursor_lim) {
-        return -1;
-    }
-    if (*cursor == ' ') {
-        cursor++;
-        goto _loop1;
-    }
-
-    // uri 
-    char *uri_lim = strchr(cursor, ' ');
-    if (!uri_lim) {
-        return -1;
-    }
-    if (uri_lim > cursor_lim) {
-        return -1;
-    }
-
-    diff = uri_lim-cursor;
-    titl->uri.er = cursor;
-    titl->uri.len = diff;
-
-    cursor += diff;
-    
-    // white space 
-_loop2:
-    if (cursor > cursor_lim) {
-        return -1;
-    }
-    if (*cursor == ' ') {
-        cursor++;
-        goto _loop2;
-    }
-
-    // ver
-    diff = cursor_lim-cursor;
-    titl->ver.er = cursor;
-    titl->ver.len = diff;
-
-    cursor += diff;
-
-    return 0;
-}
-
-int parsme(char *buff, struct httpars *pars) {
+int pareq(char *buff, struct httpareq *req) {
     int ret; 
     char *buff_lim = buff+strlen(buff);
 
     char *title_offset = buff;
     int title_len = readlin(&buff, &buff_lim);
-    if ((ret = parstitle(title_offset, title_len, &pars->titl)) < 0) {
+    if ((ret = pareqtitl(title_offset, title_len, &req->titl)) < 0) {
         fprintf(stderr, "Failed parsing title\n");
         return -1;
     }
@@ -327,7 +452,7 @@ int parsme(char *buff, struct httpars *pars) {
 
         if ((ret = parshfield(header_offset, 
                         header_len, 
-                        pars->hentries)) < 0) {
+                        req->hentries)) < 0) {
             fprintf(stderr, "Failed parsing header\n");
             return -1;
         }
@@ -336,3 +461,34 @@ int parsme(char *buff, struct httpars *pars) {
     return 0;
 }
 
+int pares(char *buff, struct httpares *res) {
+    int ret; 
+    char *buff_lim = buff+strlen(buff);
+
+    char *title_offset = buff;
+    int title_len = readlin(&buff, &buff_lim);
+    if ((ret = parestitl(title_offset, title_len, &res->titl)) < 0) {
+        fprintf(stderr, "Failed parsing title\n");
+        return -1;
+    }
+
+    for (int bound = 0; bound < MAX_BOUND; bound++) {
+        char *header_offset = buff; 
+        int header_len = readlin(&buff, &buff_lim);
+        char *header_limit = header_offset+header_len;
+        
+        // IF END OF MESSAGE
+        if (!header_len) {
+            break;
+        }
+
+        if ((ret = parshfield(header_offset, 
+                        header_len, 
+                        res->hentries)) < 0) {
+            fprintf(stderr, "Failed parsing header\n");
+            return -1;
+        }
+    }
+
+    return 0;
+}

parslib.h

@@ -9,7 +9,9 @@
                  "User-Agent: curl/8.9.1\r\n"\
                  "Accept: */*\r\n"\
                  "\r\n"\
-                 "{\"key\": \"kefjoiawejfojgorgjbosejrgo\"}"\
+                 "{\"key\": \"kefjoiawejfojgorgjbosejrgo\"}"
+#define DEFAULT_HTTP "80"
+#define DEFAULT_HTTPS "443"
 
 // SRC:https://developer.mozilla.org/en-US/docs/Web/HTTP/Methods
 enum methods {
@@ -143,6 +145,13 @@ struct point {
     int len;
 };
 
+struct hostinfo {
+    char *hostname; 
+    char *service;
+    int hostname_len;
+    int service_len;
+};
+
 struct httitlereq {
     int method;
     struct point uri; 
@@ -163,14 +172,24 @@ struct httpareq {
 struct httpares {
     struct httitleres titl; 
     struct point hentries[header_count];
-}
+};
 
+int stoin(char *str, int len, int *out);
 int initres(void);
 void fretres(void);
+void frepareq(struct httpareq *req);
+void frepares(struct httpares *res);
 int readlin(char **buff, char **buff_lim);
 int parshfield(char *offset, int len, /* out */ struct point *hentries);
-int parstitle(char *offset, int len, /* out */ struct httitle *titl);
-int parsme(char *buff, /* out */ struct httpars *pars);
-void printfpars(struct httpars *pars); 
+
+int pahostinfo(char *offset, int len, /* out */ struct hostinfo *info);
+int pareqtitl(char *offset, int len, struct httitlereq *titl);
+int parestitl(char *offset, int len, struct httitleres *titl);
+
+int pareq(char *buff, /* out */ struct httpareq *req);
+int pares(char *buff, /* out */ struct httpares *res);
+
+void printfpareq(struct httpareq *req); 
+void printfpares(struct httpares *res);
 
 #endif

streecmp

@@ -1 +1 @@
-Subproject commit ca1293aa8f123adb91d80f687933ea5c596c114a
+Subproject commit 29d80f211dde476f138f984fd7199bb6b1365034

test.c

@@ -0,0 +1,96 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "parslib.h"
+
+char *req_test_data = {
+    "GET / HTTP/1.1\r\n"
+    "Host: archive.0xdeadbeer.xyz\r\n"
+    "User-Agent: curl/8.9.1\r\n"
+    "Accept: */*\r\n"
+    "\r\n"
+    "{\"key\": \"kefjoiawejfojgorgjbosejrgo\"}|"
+    "GET / HTTP/1.1\r\n"
+    "Host: google.com\r\n"
+    "User-Agent: curl/8.9.1\r\n"
+    "Accept: */*\r\n|"
+};
+char *res_test_data = {
+    "HTTP/1.1 301 Moved Permanently\r\n"
+    "Location: http://www.google.com/\r\n"
+    "Content-Type: text/html; charset=UTF-8\r\n"
+    "Content-Security-Policy-Report-Only: object-src 'none';base-uri"
+    "'self';script-src 'nonce-7hRaJZVpUAa6E54ys6qpuA' 'strict-dynamic'"
+    "'report-sample' 'unsafe-eval' 'unsafe-inline' https:"
+    "http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp\r\n"
+    "Date: Wed, 04 Sep 2024 17:52:02 GMT\r\n"
+    "Expires: Fri, 04 Oct 2024 17:52:02 GMT\r\n"
+    "Cache-Control: public, max-age=2592000\r\n"
+    "Server: gws\r\n"
+    "Content-Length: 219\r\n"
+    "X-XSS-Protection: 0\r\n"
+    "X-Frame-Options: SAMEORIGIN\r\n|"
+};
+
+int main(void) {
+    int ret = 0;
+    char *reqs = strdup(req_test_data);
+    if (!reqs) {
+        fprintf(stderr, "Not enough dynamic memory\n");
+        return -1;
+    }
+
+    char *reses = strdup(res_test_data);
+    if (!reses) {
+        fprintf(stderr, "Not enough dynamic memory\n");
+        return -1;
+    }
+
+    struct httpareq *req = 
+        (struct httpareq *) calloc(1, sizeof(struct httpareq));
+    if (!req) {
+        fprintf(stderr, "Not enough dynamic memory\n");
+        return -1;
+    }
+
+    struct httpares *res = 
+        (struct httpares *) calloc(1, sizeof(struct httpares));
+    if (!res) {
+        fprintf(stderr, "Not enough dynamic memory\n");
+        return -1;
+    }
+
+    ret = initres(); 
+    if (ret < 0) {
+        fprintf(stderr, "Failed initializing parser\n");
+        return -1;
+    }
+
+    fprintf(stdout, "xxxREQxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx\n");
+    for (char *ch = strtok(reqs, "|"); ch != NULL; ch = strtok(NULL, "|")) {
+        ret = pareq(ch, req);
+        fprintf(stdout, "-------------------------------------\n");
+        printfpareq(req);
+        fprintf(stdout, "-------------------------------------\n");
+
+        frepareq(req);
+    }
+    fprintf(stdout, "xxxRESxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx\n");
+    for (char *ch = strtok(reses, "|"); ch != NULL; ch = strtok(NULL, "|")) {
+        ret = pares(ch, res);
+        fprintf(stdout, "-------------------------------------\n");
+        printfpares(res);
+        fprintf(stdout, "-------------------------------------\n");
+
+        frepares(res);
+    }
+    fprintf(stdout, "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx\n");
+
+    fretres();
+    free(res);
+    free(req);
+    free(reses);
+    free(reqs);
+
+    return 0;
+}