Rev 9408 | Details | Compare with Previous | Last modification | View Log | RSS feed
| Rev | Author | Line No. | Line |
|---|---|---|---|
| 10051 | ace_dent | 1 | import re |
| 2 | import os |
||
| 3 | import argparse |
||
| 4 | import sys |
||
| 5 | import pickle |
||
| 6 | import hashlib |
||
| 7 | import difflib |
||
| 8 | |||
| 9 | # fasm keywords |
||
| 10 | keywords = [ |
||
| 11 | "align", "equ", "org", "while", "load", "store", "times", "repeat", |
||
| 12 | "display", "err", "assert", "if", "aaa", "aad", "aam", "aas", "adc", |
||
| 13 | "add", "addpd", "addps", "addsd", "addss", "addsubpd", "addsubps", "adox", |
||
| 14 | "aesdeclast", "aesenc", "aesenclast", "aesimc", "aeskeygenassist", "and", |
||
| 15 | "andnpd", "andnps", "andpd", "andps", "arpl", "bextr", "blendpd", |
||
| 16 | "blendvpd", "blendvps", "blsi", "blsmsk", "blsr", "bndcl", "bndcn", |
||
| 17 | "bndldx", "bndmk", "bndmov", "bndstx", "bound", "bsf", "bsr", "bswap", |
||
| 18 | "btc", "btr", "bts", "bzhi", "call", "cbw", "cdq", "cdqe", "clac", "clc", |
||
| 19 | "cldemote", "clflush", "clflushopt", "cli", "clts", "clwb", "cmc", "cmova", |
||
| 20 | "cmovb", "cmovbe", "cmovc", "cmove", "cmovg", "cmovge", "cmovl", "cmovle", |
||
| 21 | "cmovnae", "cmovnb", "cmovnbe", "cmovnc", "cmovne", "cmovng", "cmovnge", |
||
| 22 | "cmovnle", "cmovno", "cmovnp", "cmovns", "cmovnz", "cmovo", "cmovp", |
||
| 23 | "cmovpo", "cmovs", "cmovz", "cmp", "cmppd", "cmpps", "cmps", "cmpsb", |
||
| 24 | "cmpsd", "cmpsq", "cmpss", "cmpsw", "cmpxchg", "cmpxchg16b", "cmpxchg8b", |
||
| 25 | "comiss", "cpuid", "cqo", "crc32", "cvtdq2pd", "cvtdq2ps", "cvtpd2dq", |
||
| 26 | "cvtpd2ps", "cvtpi2pd", "cvtpi2ps", "cvtps2dq", "cvtps2pd", "cvtps2pi", |
||
| 27 | "cvtsd2ss", "cvtsi2sd", "cvtsi2ss", "cvtss2sd", "cvtss2si", "cvttpd2dq", |
||
| 28 | "cvttps2dq", "cvttps2pi", "cvttsd2si", "cvttss2si", "cwd", "cwde", "daa", |
||
| 29 | "dec", "div", "divpd", "divps", "divsd", "divss", "dppd", "dpps", "emms", |
||
| 30 | "extractps", "f2xm1", "fabs", "fadd", "faddp", "fbld", "fbstp", "fchs", |
||
| 31 | "fcmova", "fcmovae", "fcmovb", "fcmovbe", "fcmovc", "fcmove", "fcmovg", |
||
| 32 | "fcmovl", "fcmovle", "fcmovna", "fcmovnae", "fcmovnb", "fcmovnbe", |
||
| 33 | "fcmovne", "fcmovng", "fcmovnge", "fcmovnl", "fcmovnle", "fcmovno", |
||
| 34 | "fcmovns", "fcmovnz", "fcmovo", "fcmovp", "fcmovpe", "fcmovpo", "fcmovs", |
||
| 35 | "fcom", "fcomi", "fcomip", "fcomp", "fcompp", "fcos", "fdecstp", "fdiv", |
||
| 36 | "fdivr", "fdivrp", "ffree", "fiadd", "ficom", "ficomp", "fidiv", "fidivr", |
||
| 37 | "fimul", "fincstp", "finit", "fist", "fistp", "fisttp", "fisub", "fisubr", |
||
| 38 | "fld1", "fldcw", "fldenv", "fldl2e", "fldl2t", "fldlg2", "fldln2", "fldpi", |
||
| 39 | "fmul", "fmulp", "fnclex", "fninit", "fnop", "fnsave", "fnstcw", "fnstenv", |
||
| 40 | "fpatan", "fprem", "fprem1", "fptan", "frndint", "frstor", "fsave", |
||
| 41 | "fsin", "fsincos", "fsqrt", "fst", "fstcw", "fstenv", "fstp", "fstsw", |
||
| 42 | "fsubp", "fsubr", "fsubrp", "ftst", "fucom", "fucomi", "fucomip", "fucomp", |
||
| 43 | "fwait", "fxam", "fxch", "fxrstor", "fxsave", "fxtract", "fyl2x", |
||
| 44 | "gf2p8affineinvqb", "gf2p8affineqb", "gf2p8mulb", "haddpd", "haddps", |
||
| 45 | "hsubpd", "hsubps", "idiv", "imul", "in", "inc", "ins", "insb", "insd", |
||
| 46 | "insw", "int", "int1", "int3", "into", "invd", "invlpg", "invpcid", "iret", |
||
| 47 | "jmp", "ja", "jae", "jb", "jbe", "jc", "jcxz", "jecxz", "je", "jg", "jge", |
||
| 48 | "jle", "jna", "jnae", "jnb", "jnbe", "jnc", "jne", "jng", "jnge", "jnl", |
||
| 49 | "jno", "jnp", "jns", "jnz", "jo", "jp", "jpe", "jpo", "js", "jz", "kaddb", |
||
| 50 | "kaddq", "kaddw", "kandb", "kandd", "kandnb", "kandnd", "kandnq", "kandnw", |
||
| 51 | "kandw", "kmovb", "kmovd", "kmovq", "kmovw", "knotb", "knotd", "knotq", |
||
| 52 | "korb", "kord", "korq", "kortestb", "kortestd", "kortestq", "kortestw", |
||
| 53 | "kshiftlb", "kshiftld", "kshiftlq", "kshiftlw", "kshiftrb", "kshiftrd", |
||
| 54 | "kshiftrw", "ktestb", "ktestd", "ktestq", "ktestw", "kunpckbw", "kunpckdq", |
||
| 55 | "kxnorb", "kxnord", "kxnorq", "kxnorw", "kxorb", "kxord", "kxorq", "kxorw", |
||
| 56 | "lar", "lddqu", "ldmxcsr", "lds", "lea", "leave", "les", "lfence", "lfs", |
||
| 57 | "lgs", "lidt", "lldt", "lmsw", "lock", "lods", "lodsb", "lodsd", "lodsq", |
||
| 58 | "loop", "loopa", "loopae", "loopb", "loopbe", "loopc", "loope", "loopg", |
||
| 59 | "loopl", "loople", "loopna", "loopnae", "loopnb", "loopnbe", "loopnc", |
||
| 60 | "loopng", "loopnge", "loopnl", "loopnle", "loopno", "loopnp", "loopns", |
||
| 61 | "loopo", "loopp", "looppe", "looppo", "loops", "loopz", "lsl", "lss", |
||
| 62 | "lzcnt", "maskmovdqu", "maskmovq", "maxpd", "maxps", "maxsd", "maxss", |
||
| 63 | "minpd", "minps", "minsd", "minss", "monitor", "mov", "movapd", "movaps", |
||
| 64 | "movd", "movddup", "movdir64b", "movdiri", "movdq2q", "movdqa", "movdqu", |
||
| 65 | "movhpd", "movhps", "movlhps", "movlpd", "movlps", "movmskpd", "movmskps", |
||
| 66 | "movntdqa", "movnti", "movntpd", "movntps", "movntq", "movq", "movq", |
||
| 67 | "movs", "movsb", "movsd", "movsd", "movshdup", "movsldup", "movsq", |
||
| 68 | "movsw", "movsx", "movsxd", "movupd", "movups", "movzx", "mpsadbw", "mul", |
||
| 69 | "mulps", "mulsd", "mulss", "mulx", "mwait", "neg", "nop", "not", "or", |
||
| 70 | "orps", "out", "outs", "outsb", "outsd", "outsw", "pabsb", "pabsd", |
||
| 71 | "pabsw", "packssdw", "packsswb", "packusdw", "packuswb", "paddb", "paddd", |
||
| 72 | "paddsb", "paddsw", "paddusb", "paddusw", "paddw", "palignr", "pand", |
||
| 73 | "pause", "pavgb", "pavgw", "pblendvb", "pblendw", "pclmulqdq", "pcmpeqb", |
||
| 74 | "pcmpeqq", "pcmpeqw", "pcmpestri", "pcmpestrm", "pcmpgtb", "pcmpgtd", |
||
| 75 | "pcmpgtw", "pcmpistri", "pcmpistrm", "pdep", "pext", "pextrb", "pextrd", |
||
| 76 | "pextrw", "phaddd", "phaddsw", "phaddw", "phminposuw", "phsubd", "phsubsw", |
||
| 77 | "pinsrb", "pinsrd", "pinsrq", "pinsrw", "pmaddubsw", "pmaddwd", "pmaxsb", |
||
| 78 | "pmaxsq", "pmaxsw", "pmaxub", "pmaxud", "pmaxuq", "pmaxuw", "pminsb", |
||
| 79 | "pminsq", "pminsw", "pminub", "pminud", "pminuq", "pminuw", "pmovmskb", |
||
| 80 | "pmovzx", "pmuldq", "pmulhrsw", "pmulhuw", "pmulhw", "pmulld", "pmullq", |
||
| 81 | "pmuludq", "pop", "popa", "popad", "popcnt", "popf", "popfd", "popfq", |
||
| 82 | "prefetchw", "prefetchh", "psadbw", "pshufb", "pshufd", "pshufhw", |
||
| 83 | "pshufw", "psignb", "psignd", "psignw", "pslld", "pslldq", "psllq", |
||
| 84 | "psrad", "psraq", "psraw", "psrld", "psrldq", "psrlq", "psrlw", "psubb", |
||
| 85 | "psubq", "psubsb", "psubsw", "psubusb", "psubusw", "psubw", "ptest", |
||
| 86 | "punpckhbw", "punpckhdq", "punpckhqdq", "punpckhwd", "punpcklbw", |
||
| 87 | "punpcklqdq", "punpcklwd", "push", "pushw", "pushd", "pusha", "pushad", |
||
| 88 | "pushfd", "pushfq", "pxor", "rcl", "rcpps", "rcpss", "rcr", "rdfsbase", |
||
| 89 | "rdmsr", "rdpid", "rdpkru", "rdpmc", "rdrand", "rdseed", "rdtsc", "rdtscp", |
||
| 90 | "repe", "repne", "repnz", "repz", "ret", "rol", "ror", "rorx", "roundpd", |
||
| 91 | "roundsd", "roundss", "rsm", "rsqrtps", "rsqrtss", "sahf", "sal", "sar", |
||
| 92 | "sbb", "scas", "scasb", "scasd", "scasw", "seta", "setae", "setb", "setbe", |
||
| 93 | "sete", "setg", "setge", "setl", "setle", "setna", "setnae", "setnb", |
||
| 94 | "setnc", "setne", "setng", "setnge", "setnl", "setnle", "setno", "setnp", |
||
| 95 | "setnz", "seto", "setp", "setpe", "setpo", "sets", "setz", "sfence", |
||
| 96 | "sha1msg1", "sha1msg2", "sha1nexte", "sha1rnds4", "sha256msg1", |
||
| 97 | "sha256rnds2", "shl", "shld", "shlx", "shr", "shrd", "shrx", "shufpd", |
||
| 98 | "sidt", "sldt", "smsw", "sqrtpd", "sqrtps", "sqrtsd", "sqrtss", "stac", |
||
| 99 | "std", "sti", "stmxcsr", "stos", "stosb", "stosd", "stosq", "stosw", "str", |
||
| 100 | "subpd", "subps", "subsd", "subss", "swapgs", "syscall", "sysenter", |
||
| 101 | "sysret", "test", "tpause", "tzcnt", "ucomisd", "ucomiss", "ud", |
||
| 102 | "umwait", "unpckhpd", "unpckhps", "unpcklpd", "unpcklps", "valignd", |
||
| 103 | "vblendmpd", "vblendmps", "vbroadcast", "vcompresspd", "vcompressps", |
||
| 104 | "vcvtpd2udq", "vcvtpd2uqq", "vcvtph2ps", "vcvtps2ph", "vcvtps2qq", |
||
| 105 | "vcvtps2uqq", "vcvtqq2pd", "vcvtqq2ps", "vcvtsd2usi", "vcvtss2usi", |
||
| 106 | "vcvttpd2udq", "vcvttpd2uqq", "vcvttps2qq", "vcvttps2udq", "vcvttps2uqq", |
||
| 107 | "vcvttss2usi", "vcvtudq2pd", "vcvtudq2ps", "vcvtuqq2pd", "vcvtuqq2ps", |
||
| 108 | "vcvtusi2ss", "vdbpsadbw", "verr", "verw", "vexpandpd", "vexpandps", |
||
| 109 | "vextractf32x4", "vextractf32x8", "vextractf64x2", "vextractf64x4", |
||
| 110 | "vextracti32x4", "vextracti32x8", "vextracti64x2", "vextracti64x4", |
||
| 111 | "vfixupimmps", "vfixupimmsd", "vfixupimmss", "vfmadd132pd", "vfmadd132ps", |
||
| 112 | "vfmadd132ss", "vfmadd213pd", "vfmadd213ps", "vfmadd213sd", "vfmadd213ss", |
||
| 113 | "vfmadd231ps", "vfmadd231sd", "vfmadd231ss", "vfmaddsub132pd", |
||
| 114 | "vfmaddsub213pd", "vfmaddsub213ps", "vfmaddsub231pd", "vfmaddsub231ps", |
||
| 115 | "vfmsub132ps", "vfmsub132sd", "vfmsub132ss", "vfmsub213pd", "vfmsub213ps", |
||
| 116 | "vfmsub213ss", "vfmsub231pd", "vfmsub231ps", "vfmsub231sd", "vfmsub231ss", |
||
| 117 | "vfmsubadd132ps", "vfmsubadd213pd", "vfmsubadd213ps", "vfmsubadd231pd", |
||
| 118 | "vfnmadd132pd", "vfnmadd132ps", "vfnmadd132sd", "vfnmadd132ss", |
||
| 119 | "vfnmadd213ps", "vfnmadd213sd", "vfnmadd213ss", "vfnmadd231pd", |
||
| 120 | "vfnmadd231sd", "vfnmadd231ss", "vfnmsub132pd", "vfnmsub132ps", |
||
| 121 | "vfnmsub132ss", "vfnmsub213pd", "vfnmsub213ps", "vfnmsub213sd", |
||
| 122 | "vfnmsub231pd", "vfnmsub231ps", "vfnmsub231sd", "vfnmsub231ss", |
||
| 123 | "vfpclassps", "vfpclasssd", "vfpclassss", "vgatherdpd", "vgatherdpd", |
||
| 124 | "vgatherdps", "vgatherqpd", "vgatherqpd", "vgatherqps", "vgatherqps", |
||
| 125 | "vgetexpps", "vgetexpsd", "vgetexpss", "vgetmantpd", "vgetmantps", |
||
| 126 | "vgetmantss", "vinsertf128", "vinsertf32x4", "vinsertf32x8", |
||
| 127 | "vinsertf64x4", "vinserti128", "vinserti32x4", "vinserti32x8", |
||
| 128 | "vinserti64x4", "vmaskmov", "vmovdqa32", "vmovdqa64", "vmovdqu16", |
||
| 129 | "vmovdqu64", "vmovdqu8", "vpblendd", "vpblendmb", "vpblendmd", "vpblendmq", |
||
| 130 | "vpbroadcast", "vpbroadcastb", "vpbroadcastd", "vpbroadcastm", |
||
| 131 | "vpbroadcastw", "vpcmpb", "vpcmpd", "vpcmpq", "vpcmpub", "vpcmpud", |
||
| 132 | "vpcmpuw", "vpcmpw", "vpcompressd", "vpcompressq", "vpconflictd", |
||
| 133 | "vperm2f128", "vperm2i128", "vpermb", "vpermd", "vpermi2b", "vpermi2d", |
||
| 134 | "vpermi2ps", "vpermi2q", "vpermi2w", "vpermilpd", "vpermilps", "vpermpd", |
||
| 135 | "vpermq", "vpermt2b", "vpermt2d", "vpermt2pd", "vpermt2ps", "vpermt2q", |
||
| 136 | "vpermw", "vpexpandd", "vpexpandq", "vpgatherdd", "vpgatherdd", |
||
| 137 | "vpgatherdq", "vpgatherqd", "vpgatherqd", "vpgatherqq", "vpgatherqq", |
||
| 138 | "vplzcntq", "vpmadd52huq", "vpmadd52luq", "vpmaskmov", "vpmovb2m", |
||
| 139 | "vpmovdb", "vpmovdw", "vpmovm2b", "vpmovm2d", "vpmovm2q", "vpmovm2w", |
||
| 140 | "vpmovqb", "vpmovqd", "vpmovqw", "vpmovsdb", "vpmovsdw", "vpmovsqb", |
||
| 141 | "vpmovsqw", "vpmovswb", "vpmovusdb", "vpmovusdw", "vpmovusqb", "vpmovusqd", |
||
| 142 | "vpmovuswb", "vpmovw2m", "vpmovwb", "vpmultishiftqb", "vprold", "vprolq", |
||
| 143 | "vprolvq", "vprord", "vprorq", "vprorvd", "vprorvq", "vpscatterdd", |
||
| 144 | "vpscatterqd", "vpscatterqq", "vpsllvd", "vpsllvq", "vpsllvw", "vpsravd", |
||
| 145 | "vpsravw", "vpsrlvd", "vpsrlvq", "vpsrlvw", "vpternlogd", "vpternlogq", |
||
| 146 | "vptestmd", "vptestmq", "vptestmw", "vptestnmb", "vptestnmd", "vptestnmq", |
||
| 147 | "vrangepd", "vrangeps", "vrangesd", "vrangess", "vrcp14pd", "vrcp14ps", |
||
| 148 | "vrcp14ss", "vreducepd", "vreduceps", "vreducesd", "vreducess", |
||
| 149 | "vrndscaleps", "vrndscalesd", "vrndscaless", "vrsqrt14pd", "vrsqrt14ps", |
||
| 150 | "vrsqrt14ss", "vscalefpd", "vscalefps", "vscalefsd", "vscalefss", |
||
| 151 | "vscatterdps", "vscatterqpd", "vscatterqps", "vshuff32x4", "vshuff64x2", |
||
| 152 | "vshufi64x2", "vtestpd", "vtestps", "vzeroall", "vzeroupper", "wait", |
||
| 153 | "wrfsbase", "wrgsbase", "wrmsr", "wrpkru", "xabort", "xacquire", "xadd", |
||
| 154 | "xchg", "xend", "xgetbv", "xlat", "xlatb", "xor", "xorpd", "xorps", |
||
| 155 | "xrstor", "xrstors", "xsave", "xsavec", "xsaveopt", "xsaves", "xsetbv", |
||
| 156 | ] |
||
| 157 | |||
| 158 | fasm_types = [ |
||
| 159 | "db", "rb", |
||
| 160 | "dw", "rw", |
||
| 161 | "dd", "rd", |
||
| 162 | "dp", "rp", |
||
| 163 | "df", "rf", |
||
| 164 | "dq", "rq", |
||
| 165 | "dt", "rt", |
||
| 166 | "du", |
||
| 167 | ] |
||
| 168 | |||
| 169 | |||
| 170 | # Add kind flag to identifier in id2kind |
||
| 171 | def id_add_kind(identifier, kind): |
||
| 172 | if identifier not in id2kind: |
||
| 173 | id2kind[identifier] = '' |
||
| 174 | id2kind[identifier] += kind |
||
| 175 | |||
| 176 | |||
| 177 | # Remove kind flag of identifier in id2kind |
||
| 178 | def id_remove_kind(identifier, kind): |
||
| 179 | if identifier in id2kind: |
||
| 180 | if kind in id2kind[identifier]: |
||
| 181 | id2kind[identifier] = id2kind[identifier].replace(kind, '') |
||
| 182 | |||
| 183 | |||
| 184 | # Get kind of an identifier |
||
| 185 | def id_get_kind(identifier): |
||
| 186 | if identifier in id2kind: |
||
| 187 | return id2kind[identifier] |
||
| 188 | else: |
||
| 189 | return '' |
||
| 190 | |||
| 191 | |||
| 192 | class LegacyAsmReader: |
||
| 193 | def __init__(self, file): |
||
| 194 | self.file = file |
||
| 195 | self.lines = open(file, "r", encoding="utf-8").readlines() |
||
| 196 | self.line_idx = 0 |
||
| 197 | self.i = 0 |
||
| 198 | |||
| 199 | def currline(self): |
||
| 200 | return self.lines[self.line_idx] |
||
| 201 | |||
| 202 | def curr(self): |
||
| 203 | try: |
||
| 204 | return self.lines[self.line_idx][self.i] |
||
| 205 | except: |
||
| 206 | return '' |
||
| 207 | |||
| 208 | def step(self): |
||
| 209 | c = self.curr() |
||
| 210 | self.i += 1 |
||
| 211 | # Wrap the line if '\\' followed by whitespaces and/or comment |
||
| 212 | while self.curr() == '\\': |
||
| 213 | i_of_backslash = self.i |
||
| 214 | self.i += 1 |
||
| 215 | while self.curr().isspace(): |
||
| 216 | self.i += 1 |
||
| 217 | if self.curr() == ';' or self.curr() == '': |
||
| 218 | self.line_idx += 1 |
||
| 219 | self.i = 0 |
||
| 220 | else: |
||
| 221 | # There's something other than a comment after the backslash |
||
| 222 | # So don't interpret the backslash as a line wrap |
||
| 223 | self.i = i_of_backslash |
||
| 224 | break |
||
| 225 | return c |
||
| 226 | |||
| 227 | def nextline(self): |
||
| 228 | c = self.curr() |
||
| 229 | while c != '': |
||
| 230 | c = self.step() |
||
| 231 | self.line_idx += 1 |
||
| 232 | self.i = 0 |
||
| 233 | |||
| 234 | def no_lines(self): |
||
| 235 | if self.line_idx >= len(self.lines): |
||
| 236 | return True |
||
| 237 | return False |
||
| 238 | |||
| 239 | def location(self): |
||
| 240 | return f"{self.file}:{self.line_idx + 1}" |
||
| 241 | |||
| 242 | def skip_spaces(self): |
||
| 243 | while self.curr().isspace(): |
||
| 244 | self.step() |
||
| 245 | |||
| 246 | |||
| 247 | class AsmReaderRecognizingStrings(LegacyAsmReader): |
||
| 248 | def __init__(self, file): |
||
| 249 | super().__init__(file) |
||
| 250 | self.in_string = None |
||
| 251 | self.should_recognize_strings = True |
||
| 252 | |||
| 253 | def step(self): |
||
| 254 | c = super().step() |
||
| 255 | if self.should_recognize_strings and (c == '"' or c == "'"): |
||
| 256 | # If just now we was at the double or single quotation mark |
||
| 257 | # and we aren't in a string yet then say |
||
| 258 | # "we are in a string openned with this quotation mark now" |
||
| 259 | if self.in_string is None: |
||
| 260 | self.in_string = c |
||
| 261 | # If just now we was at the double or single quotation mark |
||
| 262 | # and we are in the string entered with the same quotation mark |
||
| 263 | # then say "we aren't in a string anymore" |
||
| 264 | elif self.in_string == c: |
||
| 265 | self.in_string = None |
||
| 266 | return c |
||
| 267 | |||
| 268 | |||
| 269 | class AsmReaderReadingComments(AsmReaderRecognizingStrings): |
||
| 270 | def __init__(self, file): |
||
| 271 | super().__init__(file) |
||
| 272 | self.status = dict() |
||
| 273 | self.status_reset() |
||
| 274 | self.comment = '' |
||
| 275 | |||
| 276 | def status_reset(self): |
||
| 277 | # If the line has non-comment code |
||
| 278 | self.status_has_code = False |
||
| 279 | # If the line has a comment at the end |
||
| 280 | self.status_has_comment = False |
||
| 281 | # Let it recognize strings further, we are definitely out of a comment |
||
| 282 | self.should_recognize_strings = True |
||
| 283 | |||
| 284 | def status_set_has_comment(self): |
||
| 285 | self.status_has_comment = True |
||
| 286 | # Don't let it recognize strings cause we are in a comment now |
||
| 287 | self.should_recognize_strings = False |
||
| 288 | |||
| 289 | def status_set_has_code(self): |
||
| 290 | self.status_has_code = True |
||
| 291 | |||
| 292 | def update_status(self): |
||
| 293 | # If we aren't in a comment and we aren't in a string - |
||
| 294 | # say we are now in a comment if ';' met |
||
| 295 | if (not self.status_has_comment and |
||
| 296 | not self.in_string and |
||
| 297 | self.curr() == ';'): |
||
| 298 | self.status_set_has_comment() |
||
| 299 | # Else if we are in a comment - collect the comment |
||
| 300 | elif self.status_has_comment: |
||
| 301 | self.comment += self.curr() |
||
| 302 | # Else if there's some non-whitespace character out of a comment |
||
| 303 | # then the line has code |
||
| 304 | elif not self.status_has_comment and not self.curr().isspace(): |
||
| 305 | self.status_set_has_code() |
||
| 306 | |||
| 307 | def step(self): |
||
| 308 | # Get to the next character |
||
| 309 | c = super().step() |
||
| 310 | # Update status of the line according to the next character |
||
| 311 | self.update_status() |
||
| 312 | return c |
||
| 313 | |||
| 314 | def nextline(self): |
||
| 315 | prev_line = self.currline() |
||
| 316 | super().nextline() |
||
| 317 | # If the line we leave was not a comment-only line |
||
| 318 | # then forget the collected comment |
||
| 319 | # Otherwise the collected comment should be complemented by |
||
| 320 | # comment from next line in step() |
||
| 321 | if self.status_has_code: |
||
| 322 | # But we should preserve comment for the next line |
||
| 323 | # If previous line set align (cause many functions re documented |
||
| 324 | # right before align set, not before their labels) |
||
| 325 | if not prev_line.startswith("align "): |
||
| 326 | self.comment = '' |
||
| 327 | # Reset the line status (now it's the status of the new line) |
||
| 328 | self.status_reset() |
||
| 329 | # Set new status for this line according to the |
||
| 330 | # first character in the line |
||
| 331 | self.update_status() |
||
| 332 | |||
| 333 | |||
| 334 | class AsmReaderFetchingIdentifiers(AsmReaderReadingComments): |
||
| 335 | def __init__(self, file): |
||
| 336 | super().__init__(file) |
||
| 337 | |||
| 338 | def fetch_identifier(self): |
||
| 339 | self.skip_spaces() |
||
| 340 | result = '' |
||
| 341 | while is_id(self.curr()): |
||
| 342 | result += self.step() |
||
| 343 | return result |
||
| 344 | |||
| 345 | |||
| 346 | class AsmReader(AsmReaderFetchingIdentifiers): |
||
| 347 | def __init__(self, file): |
||
| 348 | super().__init__(file) |
||
| 349 | |||
| 350 | |||
| 351 | def append_file(full_path, contents): |
||
| 352 | if debug_mode: |
||
| 353 | if full_path not in output_files: |
||
| 354 | output_files[full_path] = "" |
||
| 355 | output_files[full_path] += contents |
||
| 356 | else: |
||
| 357 | f = open(full_path, "a") |
||
| 358 | f.write(contents) |
||
| 359 | f.close() |
||
| 360 | |||
| 361 | |||
| 362 | class AsmElement: |
||
| 363 | def __init__(self, location, name, comment): |
||
| 364 | global warnings |
||
| 365 | |||
| 366 | # If the element was constructed during this execution then |
||
| 367 | # the element is new |
||
| 368 | self.new = True |
||
| 369 | self.location = location |
||
| 370 | self.file = self.location.split(':')[0].replace('\\', '/') |
||
| 371 | self.line = self.location.split(':')[1] |
||
| 372 | self.name = name |
||
| 373 | self.comment = comment |
||
| 374 | |||
| 375 | if self.comment == '': |
||
| 376 | warnings += f'{self.location}: Undocumented element\n' |
||
| 377 | |||
| 378 | def dump(self): |
||
| 379 | print(f"\n{self.location}: {self.name}") |
||
| 380 | print(f"{self.comment}") |
||
| 381 | |||
| 382 | def emit(self, dest, doxycomment='', declaration=''): |
||
| 383 | # Do not emit anything if the symbol is marked as hidden in its comment |
||
| 384 | if '@dont_give_a_doxygen' in self.comment: |
||
| 385 | return |
||
| 386 | |||
| 387 | global warnings |
||
| 388 | # Redefine default declaration |
||
| 389 | if declaration == '': |
||
| 390 | declaration = f'#define {self.name}' |
||
| 391 | # Check doxycomment |
||
| 392 | if not doxycomment.endswith('\n'): |
||
| 393 | doxycomment += '\n' |
||
| 394 | if doxycomment.split('@brief ')[1][0].islower(): |
||
| 395 | warnings += (f"{self.location}: Brief comment starting from " + |
||
| 396 | "lowercase\n") |
||
| 397 | # Build contents to emit |
||
| 398 | contents = '' |
||
| 399 | contents += '/**\n' |
||
| 400 | contents += doxycomment |
||
| 401 | contents += (f"@par Source\n" + |
||
| 402 | f"{self.file}:{self.line}\n") |
||
| 404 | contents += '*/\n' |
||
| 405 | contents += declaration |
||
| 406 | contents += '\n\n' |
||
| 407 | # Get path to file to emit this |
||
| 408 | full_path = dest + '/' + self.file |
||
| 409 | # Remove the file on first access if it was |
||
| 410 | # created by previous generation |
||
| 411 | if full_path not in created_files: |
||
| 412 | if os.path.isfile(full_path): |
||
| 413 | os.remove(full_path) |
||
| 414 | created_files.append(full_path) |
||
| 415 | # Create directories need for the file |
||
| 416 | os.makedirs(os.path.dirname(full_path), exist_ok=True) |
||
| 417 | contents = ''.join([i if ord(i) < 128 else '?' for i in contents]) |
||
| 418 | |||
| 419 | append_file(full_path, contents) |
||
| 420 | |||
| 421 | |||
| 422 | class AsmVariable(AsmElement): |
||
| 423 | def __init__(self, location, name, comment, type, init): |
||
| 424 | super().__init__(location, name, comment) |
||
| 425 | self.type = type |
||
| 426 | self.init = init |
||
| 427 | |||
| 428 | def dump(self): |
||
| 429 | super().dump() |
||
| 430 | print(f"(Variable)\n---") |
||
| 431 | |||
| 432 | def emit(self, dest): |
||
| 433 | # Build doxycomment specific for the variable |
||
| 434 | doxycomment = '' |
||
| 435 | doxycomment += self.comment |
||
| 436 | if '@brief' not in doxycomment: |
||
| 437 | doxycomment = '@brief ' + doxycomment |
||
| 438 | doxycomment += (f"@par Initial value\n" + |
||
| 439 | f"{self.init}\n") |
||
| 440 | # Build the declaration |
||
| 441 | name = self.name.replace(".", "_") |
||
| 442 | var_type = self.type.replace(".", "_") |
||
| 443 | declaration = f"{var_type} {name};" |
||
| 444 | # Emit this |
||
| 445 | super().emit(dest, doxycomment, declaration) |
||
| 446 | |||
| 447 | |||
| 448 | class AsmFunction(AsmElement): |
||
| 449 | def __init__(self, location, name, comment, calling_convention, |
||
| 450 | args, used_regs): |
||
| 451 | super().__init__(location, name, comment) |
||
| 452 | self.calling_convention = calling_convention |
||
| 453 | self.args = args |
||
| 454 | self.used_regs = used_regs |
||
| 455 | |||
| 456 | def dump(self): |
||
| 457 | super().dump() |
||
| 458 | print(f"(Function)\n---") |
||
| 459 | |||
| 460 | def emit(self, dest): |
||
| 461 | # Build doxycomment specific for the variable |
||
| 462 | doxycomment = '' |
||
| 463 | doxycomment += self.comment |
||
| 464 | if '@brief' not in doxycomment: |
||
| 465 | doxycomment = '@brief ' + doxycomment |
||
| 466 | # If there was no arguments, maybe that's just a label |
||
| 467 | # then parse parameters from its comment |
||
| 468 | if len(self.args) == 0 and '@param' in self.comment: |
||
| 469 | i = 0 |
||
| 470 | while '@param' in self.comment[i:]: |
||
| 471 | i = self.comment.index('@param', i) |
||
| 472 | # Skip '@param' |
||
| 473 | i += len('@param') |
||
| 474 | # Skip spaces after '@param' |
||
| 475 | while self.comment[i].isspace(): |
||
| 476 | i += 1 |
||
| 477 | # Get the parameter name |
||
| 478 | name = '' |
||
| 479 | while is_id(self.comment[i]): |
||
| 480 | name += self.comment[i] |
||
| 481 | i += 1 |
||
| 482 | # Save the parameter |
||
| 483 | self.args.append((name, 'arg_t')) |
||
| 484 | # Build the arg list for declaration |
||
| 485 | arg_list = '(' |
||
| 486 | if len(self.args) > 0: |
||
| 487 | argc = 0 |
||
| 488 | for arg in self.args: |
||
| 489 | if argc != 0: |
||
| 490 | arg_list += ", " |
||
| 491 | arg_list += f"{arg[1]} {arg[0]}" |
||
| 492 | argc += 1 |
||
| 493 | arg_list += ')' |
||
| 494 | # Build the declaration |
||
| 495 | name = self.name.replace(".", "_") |
||
| 496 | declaration = f"void {name}{arg_list};" |
||
| 497 | # Emit this |
||
| 498 | super().emit(dest, doxycomment, declaration) |
||
| 499 | |||
| 500 | |||
| 501 | class AsmLabel(AsmElement): |
||
| 502 | def __init__(self, location, name, comment): |
||
| 503 | super().__init__(location, name, comment) |
||
| 504 | |||
| 505 | def dump(self): |
||
| 506 | super().dump() |
||
| 507 | print(f"(Label)\n---") |
||
| 508 | |||
| 509 | def emit(self, dest): |
||
| 510 | # Build doxycomment specific for the variable |
||
| 511 | doxycomment = '' |
||
| 512 | doxycomment += self.comment |
||
| 513 | if '@brief' not in doxycomment: |
||
| 514 | doxycomment = '@brief ' + doxycomment |
||
| 515 | # Build the declaration |
||
| 516 | name = self.name.replace(".", "_") |
||
| 517 | declaration = f"label {name};" |
||
| 518 | # Emit this |
||
| 519 | super().emit(dest, doxycomment, declaration) |
||
| 520 | |||
| 521 | |||
| 522 | class AsmMacro(AsmElement): |
||
| 523 | def __init__(self, location, name, comment, args): |
||
| 524 | super().__init__(location, name, comment) |
||
| 525 | self.args = args |
||
| 526 | |||
| 527 | def dump(self): |
||
| 528 | super().dump() |
||
| 529 | print(f"(Macro)\n---") |
||
| 530 | |||
| 531 | def emit(self, dest): |
||
| 532 | # Construct arg list without '['s, ']'s and '*'s |
||
| 533 | args = [arg for arg in self.args if arg not in "[]*"] |
||
| 534 | # Construct C-like arg list |
||
| 535 | arg_list = "" |
||
| 536 | if len(args) > 0: |
||
| 537 | arg_list += '(' |
||
| 538 | argc = 0 |
||
| 539 | for arg in args: |
||
| 540 | if argc != 0: |
||
| 541 | arg_list += ", " |
||
| 542 | arg_list += arg |
||
| 543 | argc += 1 |
||
| 544 | arg_list += ')' |
||
| 545 | # Build doxycomment |
||
| 546 | doxycomment = '' |
||
| 547 | doxycomment += self.comment |
||
| 548 | if '@brief' not in doxycomment: |
||
| 549 | doxycomment = '@brief ' + doxycomment |
||
| 550 | # Build declaration |
||
| 551 | declaration = f"#define {self.name}{arg_list}" |
||
| 552 | # Emit this |
||
| 553 | super().emit(dest, doxycomment, declaration) |
||
| 554 | |||
| 555 | |||
| 556 | class AsmStruct(AsmElement): |
||
| 557 | def __init__(self, location, name, comment, members): |
||
| 558 | super().__init__(location, name, comment) |
||
| 559 | self.members = members |
||
| 560 | |||
| 561 | def dump(self): |
||
| 562 | super().dump() |
||
| 563 | print(f"(Struct)\n---") |
||
| 564 | |||
| 565 | def emit(self, dest): |
||
| 566 | # Build doxycomment |
||
| 567 | doxycomment = '' |
||
| 568 | doxycomment += self.comment |
||
| 569 | if '@brief' not in doxycomment: |
||
| 570 | doxycomment = '@brief ' + doxycomment |
||
| 571 | doxycomment += '\n' |
||
| 572 | # Build declaration |
||
| 573 | declaration = f"struct {self.name}" + " {\n" |
||
| 574 | for member in self.members: |
||
| 575 | if type(member) == AsmVariable: |
||
| 576 | declaration += (f'\t{member.type} {member.name}; ' + |
||
| 577 | f'/**< {member.comment} */\n') |
||
| 578 | declaration += '};' |
||
| 579 | # Emit this |
||
| 580 | super().emit(dest, doxycomment, declaration) |
||
| 581 | |||
| 582 | |||
| 583 | class AsmUnion(AsmElement): |
||
| 584 | def __init__(self, location, name, comment, members): |
||
| 585 | super().__init__(location, name, comment) |
||
| 586 | self.members = members |
||
| 587 | |||
| 588 | def dump(self): |
||
| 589 | super().dump() |
||
| 590 | print(f"(Union)\n---") |
||
| 591 | |||
| 592 | def emit(self, dest): |
||
| 593 | # Build doxycomment |
||
| 594 | doxycomment = '' |
||
| 595 | doxycomment += self.comment |
||
| 596 | if '@brief' not in doxycomment: |
||
| 597 | doxycomment = '@brief ' + doxycomment |
||
| 598 | # Build declaration |
||
| 599 | declaration = f"union {self.name}" + " {};" |
||
| 600 | # Emit this |
||
| 601 | super().emit(dest, doxycomment, declaration) |
||
| 602 | |||
| 603 | |||
| 604 | class VariableNameIsMacroName: |
||
| 605 | def __init__(self, name): |
||
| 606 | self.name = name |
||
| 607 | |||
| 608 | |||
| 609 | def is_id(c): |
||
| 610 | return c.isprintable() and c not in "+-/*=<>()[]{};:,|&~#`'\" \n\r\t\v" |
||
| 611 | |||
| 612 | |||
| 613 | def is_starts_as_id(s): |
||
| 614 | return not s[0].isdigit() |
||
| 615 | |||
| 616 | |||
| 617 | def parse_after_macro(r): |
||
| 618 | location = r.location() |
||
| 619 | |||
| 620 | # Skip spaces after the "macro" keyword |
||
| 621 | r.skip_spaces() |
||
| 622 | # Read macro name |
||
| 623 | name = "" |
||
| 624 | while is_id(r.curr()) or r.curr() == '#': |
||
| 625 | name += r.step() |
||
| 626 | # Skip spaces after macro name |
||
| 627 | r.skip_spaces() |
||
| 628 | # Find all arguments |
||
| 629 | args = [] |
||
| 630 | arg = '' |
||
| 631 | while r.curr() and r.curr() != ';' and r.curr() != '{': |
||
| 632 | # Collect identifier |
||
| 633 | if is_id(r.curr()): |
||
| 634 | arg += r.step() |
||
| 635 | # Save the collected identifier |
||
| 636 | elif r.curr() == ',': |
||
| 637 | args.append(arg) |
||
| 638 | arg = '' |
||
| 639 | r.step() |
||
| 640 | # Just push the '[' |
||
| 641 | elif r.curr() == '[': |
||
| 642 | args.append(r.step()) |
||
| 643 | # Just push the identifier and get ']' ready to be pushed on next comma |
||
| 644 | elif r.curr() == ']': |
||
| 645 | args.append(arg) |
||
| 646 | arg = r.step() |
||
| 647 | # Just push the identifier and get '*' ready to be pushed on next comma |
||
| 648 | elif r.curr() == '*': |
||
| 649 | args.append(arg) |
||
| 650 | arg = r.step() |
||
| 651 | # Just skip whitespaces |
||
| 652 | elif r.curr().isspace(): |
||
| 653 | r.step() |
||
| 654 | # Something unexpected |
||
| 655 | else: |
||
| 656 | raise Exception(f"Unexpected symbol '{r.curr()}' " + |
||
| 657 | f"at index #{r.i} in the macro declaration " + |
||
| 658 | f"at {location} " + |
||
| 659 | f"(line: {r.lines[r.line_idx]})\n''") |
||
| 660 | # Append the last argument |
||
| 661 | if arg != '': |
||
| 662 | args.append(arg) |
||
| 663 | # Skip t spaces after the argument list |
||
| 664 | r.skip_spaces() |
||
| 665 | # Get a comment if it is: read till the end of the line and |
||
| 666 | # get the comment from the reader |
||
| 667 | while r.curr() != '': |
||
| 668 | r.step() |
||
| 669 | comment = r.comment |
||
| 670 | # Find end of the macro |
||
| 671 | prev = '' |
||
| 672 | while True: |
||
| 673 | if r.curr() == '}' and prev != '\\': |
||
| 674 | break |
||
| 675 | elif r.curr() == '': |
||
| 676 | prev = '' |
||
| 677 | r.nextline() |
||
| 678 | continue |
||
| 679 | prev = r.step() |
||
| 680 | # Build the output |
||
| 681 | return AsmMacro(location, name, comment, args) |
||
| 682 | |||
| 683 | |||
| 684 | def parse_variable(r, first_word=None): |
||
| 685 | global warnings |
||
| 686 | location = r.location() |
||
| 687 | |||
| 688 | # Skip spaces before variable name |
||
| 689 | r.skip_spaces() |
||
| 690 | # Get variable name |
||
| 691 | name = "" |
||
| 692 | # Read it if it was not supplied |
||
| 693 | if first_word is None: |
||
| 694 | while is_id(r.curr()): |
||
| 695 | name += r.step() |
||
| 696 | # Or use the supplied one instead |
||
| 697 | else: |
||
| 698 | name = first_word |
||
| 699 | # Check the name |
||
| 700 | # If it's 0 len, that means threr's something else than an |
||
| 701 | # identifier at the beginning |
||
| 702 | if len(name) == 0: |
||
| 703 | return None |
||
| 704 | # If it starts from digit or othervice illegally it's illegal |
||
| 705 | if not is_starts_as_id(name): |
||
| 706 | return None |
||
| 707 | # Get kind of the identifier from id2kind table |
||
| 708 | kind = id_get_kind(name) |
||
| 709 | # If it's a keyword, that's not a variable declaration |
||
| 710 | if ID_KIND_KEYWORD in kind: |
||
| 711 | return None |
||
| 712 | # If it's a macro name, that's not a variable declaration |
||
| 713 | if ID_KIND_MACRO_NAME in kind: |
||
| 714 | return VariableNameIsMacroName(name) |
||
| 715 | # If it's a datatype or a structure name that's not a |
||
| 716 | # variable declaration: that's just a data |
||
| 717 | # don't document just a data for now |
||
| 718 | if ID_KIND_STRUCT_NAME in kind or ID_KIND_FASM_TYPE in kind: |
||
| 719 | return None |
||
| 720 | # Skip spaces before type name |
||
| 721 | r.skip_spaces() |
||
| 722 | # Read type name |
||
| 723 | var_type = "" |
||
| 724 | while is_id(r.curr()): |
||
| 725 | var_type += r.step() |
||
| 726 | # Check the type name |
||
| 727 | if len(var_type) == 0: |
||
| 728 | # If there's no type identifier after the name |
||
| 729 | # maybe the name is something meaningful for the next parser |
||
| 730 | # return it |
||
| 731 | return name |
||
| 732 | # If it starts from digit or othervice illegally it's illegal |
||
| 733 | if not is_starts_as_id(var_type): |
||
| 734 | return None |
||
| 735 | # Get kind of type identifier |
||
| 736 | type_kind = id_get_kind(var_type) |
||
| 737 | # If it's a keyword, that's not a variable declaration |
||
| 738 | # return the two words of the lexical structure |
||
| 739 | if ID_KIND_KEYWORD in type_kind: |
||
| 740 | return (name, var_type) |
||
| 741 | # Skip spaces before the value |
||
| 742 | r.skip_spaces() |
||
| 743 | # Read the value until the comment or end of the line |
||
| 744 | value = "" |
||
| 745 | while r.curr() != ';' and r.curr() != '' and r.curr() != '\n': |
||
| 746 | value += r.step() |
||
| 747 | # Skip spaces after the value |
||
| 748 | r.skip_spaces() |
||
| 749 | # Read till end of the line to get a comment from the reader |
||
| 750 | while r.curr() != '': |
||
| 751 | r.step() |
||
| 752 | # Build the result |
||
| 753 | return AsmVariable(location, name, r.comment, var_type, value) |
||
| 754 | |||
| 755 | |||
| 756 | def parse_after_struct(r, as_union=True): |
||
| 757 | global warnings |
||
| 758 | location = r.location() |
||
| 759 | |||
| 760 | # Skip spaces after "struct" keyword |
||
| 761 | r.skip_spaces() |
||
| 762 | # Read struct name |
||
| 763 | name = "" |
||
| 764 | while is_id(r.curr()): |
||
| 765 | name += r.step() |
||
| 766 | # Read till end of the line and get the comment from the reader |
||
| 767 | while r.curr() != '': |
||
| 768 | r.step() |
||
| 769 | comment = r.comment |
||
| 770 | # Get to the next line to parse struct members |
||
| 771 | r.nextline() |
||
| 772 | # Parse struct members |
||
| 773 | members = [] |
||
| 774 | while True: |
||
| 775 | r.skip_spaces() |
||
| 776 | var = parse_variable(r) |
||
| 777 | if type(var) == AsmVariable: |
||
| 778 | members.append(var) |
||
| 779 | elif type(var) == str: |
||
| 780 | if var == 'union': |
||
| 781 | # Parse the union as a struct |
||
| 782 | union = parse_after_struct(r, as_union=True) |
||
| 783 | members.append(union) |
||
| 784 | # Skip the ends of the union |
||
| 785 | r.nextline() |
||
| 786 | elif r.curr() == ':': |
||
| 787 | warnings += f"{r.location()}: Skept the label in the struct\n" |
||
| 788 | else: |
||
| 789 | raise Exception(f"Garbage in struct member at {location} " + |
||
| 790 | f" (got '{var}' identifier)") |
||
| 791 | elif type(var) == VariableNameIsMacroName: |
||
| 792 | if var.name == 'ends': |
||
| 793 | break |
||
| 794 | r.nextline() |
||
| 795 | # Return the result |
||
| 796 | if as_union: |
||
| 797 | return AsmStruct(location, name, comment, members) |
||
| 798 | else: |
||
| 799 | return AsmUnion(location, name, comment, members) |
||
| 800 | |||
| 801 | |||
| 802 | def parse_after_proc(r): |
||
| 803 | # Get proc name |
||
| 804 | name = r.fetch_identifier() |
||
| 805 | # Next identifier after the proc name |
||
| 806 | identifier = r.fetch_identifier() |
||
| 807 | # Check if the id is 'stdcall' or 'c' (calling convention specifier) |
||
| 808 | # and if so - save the convention and lookup the next identifier |
||
| 809 | calling_convention = '' |
||
| 810 | if identifier == 'stdcall' or identifier == 'c': |
||
| 811 | calling_convention = identifier |
||
| 812 | # If next is a comma, just skip it |
||
| 813 | if r.curr() == ',': |
||
| 814 | r.step() |
||
| 815 | # Read the next identifier |
||
| 816 | identifier = r.fetch_identifier() |
||
| 817 | # Check if the id is 'uses' (used register list specifier) |
||
| 818 | # and if so save the used register list |
||
| 819 | used_regs = [] |
||
| 820 | if identifier == 'uses': |
||
| 821 | # Read the registers |
||
| 822 | while True: |
||
| 823 | reg_name = r.fetch_identifier() |
||
| 824 | if reg_name != '': |
||
| 825 | used_regs.append(reg_name) |
||
| 826 | else: |
||
| 827 | break |
||
| 828 | # If next is a comma, just skip it |
||
| 829 | if r.curr() == ',': |
||
| 830 | r.step() |
||
| 831 | # Read the next identifier |
||
| 832 | identifier = r.fetch_identifier() |
||
| 833 | # Check if there are argument identifiers |
||
| 834 | args = [] |
||
| 835 | while identifier != '': |
||
| 836 | arg_name = identifier |
||
| 837 | arg_type = 'arg_t' |
||
| 838 | # Skip spaces after argument name |
||
| 839 | r.skip_spaces() |
||
| 840 | # If there's a ':' after the name - the next identifier is type |
||
| 841 | if r.curr() == ':': |
||
| 842 | r.step() |
||
| 843 | arg_type = r.fetch_identifier() |
||
| 844 | # If there's a comma - there's one more argument |
||
| 845 | # else no arguments anymore |
||
| 846 | if r.curr() == ',': |
||
| 847 | r.step() |
||
| 848 | identifier = r.fetch_identifier() |
||
| 849 | else: |
||
| 850 | identifier = '' |
||
| 851 | args.append((arg_name, arg_type)) |
||
| 852 | # Get to the end of the line and get a comment from the reader |
||
| 853 | while r.curr() != '': |
||
| 854 | r.step() |
||
| 855 | comment = r.comment |
||
| 856 | # Build the element |
||
| 857 | return AsmFunction(r.location(), name, comment, calling_convention, |
||
| 858 | args, used_regs) |
||
| 859 | |||
| 860 | |||
| 861 | def get_declarations(asm_file_contents, asm_file_name): |
||
| 862 | r = AsmReader(asm_file_name) |
||
| 863 | |||
| 864 | while not r.no_lines(): |
||
| 865 | # Skip leading spaces |
||
| 866 | r.skip_spaces() |
||
| 867 | # Skip the line if it's starting with a comment |
||
| 868 | if r.curr() == ';': |
||
| 869 | r.nextline() |
||
| 870 | continue |
||
| 871 | # Get first word |
||
| 872 | first_word = "" |
||
| 873 | while is_id(r.curr()): |
||
| 874 | first_word += r.step() |
||
| 875 | # Match macro declaration |
||
| 876 | if first_word == "macro": |
||
| 877 | macro = parse_after_macro(r) |
||
| 878 | elements.append(macro) |
||
| 879 | id_add_kind(macro.name, ID_KIND_MACRO_NAME) |
||
| 880 | # Match structure declaration |
||
| 881 | elif first_word == "struct": |
||
| 882 | struct = parse_after_struct(r) |
||
| 883 | elements.append(struct) |
||
| 884 | id_add_kind(struct.name, ID_KIND_STRUCT_NAME) |
||
| 885 | # Match function definition |
||
| 886 | elif first_word == "proc": |
||
| 887 | proc = parse_after_proc(r) |
||
| 888 | elements.append(proc) |
||
| 889 | elif first_word == 'format': |
||
| 890 | # Skip the format directive |
||
| 891 | pass |
||
| 892 | elif first_word == 'include': |
||
| 893 | # Skip the include directive |
||
| 894 | pass |
||
| 895 | elif first_word == 'if': |
||
| 896 | # Skip the conditional directive |
||
| 897 | pass |
||
| 898 | elif first_word == 'repeat': |
||
| 899 | # Skip the repeat directive |
||
| 900 | pass |
||
| 901 | elif first_word == 'purge': |
||
| 902 | while True: |
||
| 903 | # Skip spaces after the 'purge' keyword or after |
||
| 904 | # the comma what separated the previous macro name |
||
| 905 | r.skip_spaces() |
||
| 906 | # Get the purged macro name |
||
| 907 | name = '' |
||
| 908 | while is_id(r.curr()): |
||
| 909 | name += r.step() |
||
| 910 | # Remove the purged macro from the macro names list |
||
| 911 | try: |
||
| 912 | id_remove_kind(name, ID_KIND_MACRO_NAME) |
||
| 913 | except: |
||
| 914 | pass |
||
| 915 | # Skip spaces after the name |
||
| 916 | r.skip_spaces() |
||
| 917 | # If it's comma (',') after then that's not the last purged |
||
| 918 | # macro, continue purging |
||
| 919 | if r.curr() == ',': |
||
| 920 | r.step() |
||
| 921 | continue |
||
| 922 | # Here we purged all the macros should be purged |
||
| 923 | break |
||
| 924 | # Match label or a variable |
||
| 925 | elif len(first_word) != 0: |
||
| 926 | # Skip spaces after the identifier |
||
| 927 | r.skip_spaces() |
||
| 928 | # Match a variable |
||
| 929 | var = parse_variable(r, first_word) |
||
| 930 | if type(var) == AsmVariable: |
||
| 931 | elements.append(var) |
||
| 932 | # If it wasn't a variable but there was an identifier |
||
| 933 | # Maybe that's a label and the identifier is the label name |
||
| 934 | # The parse_variable returns the first found or supplied identifier |
||
| 935 | # In this case it returns the first_word which is supplied |
||
| 936 | # If it didn't match a type identifier after the word |
||
| 937 | elif type(var) == str: |
||
| 938 | name = var |
||
| 939 | # Match label beginning (':' after name) |
||
| 940 | if r.curr() == ':': |
||
| 941 | # Get to the end of the line and |
||
| 942 | # get the coment from the reader |
||
| 943 | while r.curr() != '': |
||
| 944 | r.step() |
||
| 945 | comment = r.comment |
||
| 946 | # Only handle non-local labels |
||
| 947 | if name[0] != '.' and name != "@@": |
||
| 948 | # Treat the label as function if there's @return or |
||
| 949 | # @param in its comment. Otherwise it's just a variable |
||
| 950 | # with type `label` in generated doxygen C |
||
| 951 | if '@return' in comment or '@param' in comment: |
||
| 952 | element = AsmFunction(r.location(), name, comment, |
||
| 953 | '', [], []) |
||
| 954 | else: |
||
| 955 | element = AsmLabel(r.location(), name, comment) |
||
| 956 | elements.append(element) |
||
| 957 | elif r.curr() == '=': |
||
| 958 | # Save the identifier as a set constant |
||
| 959 | id_add_kind(first_word, ID_KIND_SET_CONSTANT) |
||
| 960 | elif type(var) == tuple: |
||
| 961 | (word_one, word_two) = var |
||
| 962 | if word_two == 'equ': |
||
| 963 | # Save the identifier as an equated constant |
||
| 964 | id_add_kind(word_one, ID_KIND_EQUATED_CONSTANT) |
||
| 965 | r.nextline() |
||
| 966 | |||
| 967 | |||
| 968 | def it_neds_to_be_parsed(source_file): |
||
| 969 | # If there's no symbols file saved - parse it anyway |
||
| 970 | # cause we need to create the symbols file and use it |
||
| 971 | # if we gonna generate proper doxygen |
||
| 972 | if not os.path.isfile('asmxygen.elements.pickle'): |
||
| 973 | return True |
||
| 974 | dest = doxygen_src_path + '/' + source_file |
||
| 975 | # If there's no the doxygen file it should be compiled to |
||
| 976 | # then yes, we should compile it to doxygen |
||
| 977 | if not os.path.isfile(dest): |
||
| 978 | return True |
||
| 979 | source_change_time = os.path.getmtime(source_file) |
||
| 980 | dest_change_file = os.path.getmtime(dest) |
||
| 981 | # If the source is newer than the doxygen it was compiled to |
||
| 982 | # then the source should be recompiled (existing doxygen is old) |
||
| 983 | if source_change_time > dest_change_file: |
||
| 984 | return True |
||
| 985 | return False |
||
| 986 | |||
| 987 | |||
| 988 | def handle_file(handled_files, asm_file_name, subdir="."): |
||
| 989 | global elements |
||
| 990 | # Canonicalize the file path and get it relative to cwd |
||
| 991 | cwd = os.path.abspath(os.path.dirname(sys.argv[0])) |
||
| 992 | asm_file_name = os.path.realpath(asm_file_name) |
||
| 993 | asm_file_name = asm_file_name[len(cwd) + 1:] |
||
| 994 | # If it's lang.inc - skip it |
||
| 995 | if asm_file_name == 'lang.inc': |
||
| 996 | return |
||
| 997 | # If the file was handled in this execution before - skip it |
||
| 998 | if asm_file_name in handled_files: |
||
| 999 | return |
||
| 1000 | # Say that the file was handled in this execution |
||
| 1001 | handled_files.append(asm_file_name) |
||
| 1002 | # Check if the file should be parsed |
||
| 1003 | # (if it was modified or wasn't parsed yet) |
||
| 1004 | should_get_declarations = True |
||
| 1005 | if not it_neds_to_be_parsed(asm_file_name): |
||
| 1006 | print(f"Skipping {asm_file_name} (already newest)") |
||
| 1007 | should_get_declarations = False |
||
| 1008 | else: |
||
| 1009 | print(f"Handling {asm_file_name}") |
||
| 1010 | # Remove elements parsed from this file before if any |
||
| 1011 | elements_to_remove = [ |
||
| 1012 | x for x in elements if x.location.split(':')[0] == asm_file_name |
||
| 1013 | ] |
||
| 1014 | elements = [ |
||
| 1015 | x for x in elements if x.location.split(':')[0] != asm_file_name |
||
| 1016 | ] |
||
| 1017 | # Forget types of identifiers of names of the removed elements |
||
| 1018 | for element in elements_to_remove: |
||
| 1019 | if type(element) == AsmStruct: |
||
| 1020 | id_remove_kind(element.name, ID_KIND_STRUCT_NAME) |
||
| 1021 | elif type(element) == AsmMacro: |
||
| 1022 | id_remove_kind(element.name, ID_KIND_MACRO_NAME) |
||
| 1023 | # Read the source |
||
| 1024 | asm_file_contents = open(asm_file_name, "r", encoding="utf-8").read() |
||
| 1025 | # Find includes, fix their paths and handle em recoursively |
||
| 1026 | includes = re.findall(r'^include (["\'])(.*)\1', asm_file_contents, |
||
| 1027 | flags=re.MULTILINE) |
||
| 1028 | for include in includes: |
||
| 1029 | include = include[1].replace('\\', '/') |
||
| 1030 | full_path = subdir + '/' + include |
||
| 1031 | # If the path isn't valid, maybe that's not relative path |
||
| 1032 | if not os.path.isfile(full_path): |
||
| 1033 | full_path = include |
||
| 1034 | new_subdir = full_path.rsplit('/', 1)[0] |
||
| 1035 | handle_file(handled_files, full_path, new_subdir) |
||
| 1036 | # Only collect declarations from the file if it wasn't parsed before |
||
| 1037 | if should_get_declarations and not clean_generated_stuff: |
||
| 1038 | get_declarations(asm_file_contents, asm_file_name) |
||
| 1039 | |||
| 1040 | if __name__ == "__main__": |
||
| 1041 | link_root = "http://websvn.kolibrios.org/filedetails.php" |
||
| 1042 | link_root += "?repname=Kolibri+OS&path=/kernel/trunk" |
||
| 1043 | |||
| 1044 | # Dict where an identifier is assicoated with a string |
||
| 1045 | # The string contains characters specifying flags |
||
| 1046 | # Available flags: |
||
| 1047 | # k - Keyword |
||
| 1048 | # m - Macro name |
||
| 1049 | # t - fasm data Type name (db, rq, etc.) |
||
| 1050 | # s - Struct type name |
||
| 1051 | # e - equated constant (name equ value) |
||
| 1052 | # = - set constants (name = value) |
||
| 1053 | ID_KIND_KEYWORD = 'k' |
||
| 1054 | ID_KIND_MACRO_NAME = 'm' |
||
| 1055 | ID_KIND_FASM_TYPE = 't' |
||
| 1056 | ID_KIND_STRUCT_NAME = 's' |
||
| 1057 | ID_KIND_EQUATED_CONSTANT = 'e' |
||
| 1058 | ID_KIND_SET_CONSTANT = '=' |
||
| 1059 | id2kind = {} |
||
| 1060 | |||
| 1061 | for keyword in keywords: |
||
| 1062 | id_add_kind(keyword, ID_KIND_KEYWORD) |
||
| 1063 | |||
| 1064 | for fasm_type in fasm_types: |
||
| 1065 | id_add_kind(fasm_type, ID_KIND_FASM_TYPE) |
||
| 1066 | |||
| 1067 | # Warning list |
||
| 1068 | warnings = "" |
||
| 1069 | |||
| 1070 | # Parameters |
||
| 1071 | # Path to doxygen folder to make doxygen files in: -o |
||
| 1072 | doxygen_src_path = 'docs/doxygen' |
||
| 1073 | # Remove generated doxygen files: --clean |
||
| 1074 | clean_generated_stuff = False |
||
| 1075 | # Dump all defined symbols: --dump |
||
| 1076 | dump_symbols = False |
||
| 1077 | # Print symbol stats: --stats |
||
| 1078 | print_stats = False |
||
| 1079 | # Do not write warnings file: --nowarn |
||
| 1080 | enable_warnings = True |
||
| 1081 | |||
| 1082 | # Parse arguments |
||
| 1083 | parser = argparse.ArgumentParser() |
||
| 1084 | parser.add_argument("-o", help="Doxygen output folder") |
||
| 1085 | parser.add_argument("--clean", |
||
| 1086 | help="Remove generated files", |
||
| 1087 | action="store_true") |
||
| 1088 | parser.add_argument("--dump", |
||
| 1089 | help="Dump all defined symbols", |
||
| 1090 | action="store_true") |
||
| 1091 | parser.add_argument("--stats", |
||
| 1092 | help="Print symbol stats", |
||
| 1093 | action="store_true") |
||
| 1094 | parser.add_argument("--nowarn", |
||
| 1095 | help="Do not write warnings file", |
||
| 1096 | action="store_true") |
||
| 1097 | parser.add_argument("--noemit", |
||
| 1098 | help="Do not emit doxygen files (for testing)", |
||
| 1099 | action="store_true") |
||
| 1100 | parser.add_argument("--debug", |
||
| 1101 | help="Show hashes of files (for testing)", |
||
| 1102 | action="store_true") |
||
| 1103 | args = parser.parse_args() |
||
| 1104 | doxygen_src_path = args.o if args.o else 'docs/doxygen' |
||
| 1105 | clean_generated_stuff = args.clean |
||
| 1106 | dump_symbols = args.dump |
||
| 1107 | print_stats = args.stats |
||
| 1108 | enable_warnings = not args.nowarn |
||
| 1109 | noemit = args.noemit |
||
| 1110 | debug_mode = args.debug |
||
| 1111 | |||
| 1112 | # Variables, functions, labels, macros, structure types |
||
| 1113 | elements = [] |
||
| 1114 | created_files = [] |
||
| 1115 | kernel_files = [] |
||
| 1116 | output_files = {} # If --debug then all the files are written here |
||
| 1117 | |||
| 1118 | # Load remembered list of symbols |
||
| 1119 | if os.path.isfile('asmxygen.elements.pickle'): |
||
| 1120 | print('Reading existing dump of symbols') |
||
| 1121 | pickle_file = open('asmxygen.elements.pickle', 'rb') |
||
| 1122 | (elements, id2kind) = pickle.load(pickle_file) |
||
| 1123 | pickle_file.close() |
||
| 1124 | |||
| 1125 | handle_file(kernel_files, "./kernel.asm") |
||
| 1126 | |||
| 1127 | if dump_symbols: |
||
| 1128 | stdout = sys.stdout |
||
| 1129 | sys.stdout = open('asmxygen.dump.txt', 'w', encoding='utf-8') |
||
| 1130 | for asm_element in elements: |
||
| 1131 | asm_element.dump() |
||
| 1132 | sys.stdout = stdout |
||
| 1133 | |||
| 1134 | if clean_generated_stuff: |
||
| 1135 | kernel_files_set = set(kernel_files) |
||
| 1136 | for file in kernel_files: |
||
| 1137 | doxygen_file = f"{doxygen_src_path}/{file}" |
||
| 1138 | if (os.path.isfile(doxygen_file)): |
||
| 1139 | print(f"Removing {file}... ", end='') |
||
| 1140 | os.remove(doxygen_file) |
||
| 1141 | print("Done.") |
||
| 1142 | elif not noemit: |
||
| 1143 | print(f"Writing doumented sources to {doxygen_src_path}") |
||
| 1144 | |||
| 1145 | i = 0 |
||
| 1146 | new_elements = [x for x in elements if x.new] |
||
| 1147 | for element in new_elements: |
||
| 1148 | counter = f"[{i + 1}/{len(new_elements)}]" |
||
| 1149 | print(f"{counter} Emitting {element.name} from {element.location}") |
||
| 1150 | element.emit(doxygen_src_path) |
||
| 1151 | i += 1 |
||
| 1152 | |||
| 1153 | print(f"Writing dump of symbols to asmxygen.elements.pickle") |
||
| 1154 | |||
| 1155 | # Now when the new elements already was written, there's no new |
||
| 1156 | # elements anymore |
||
| 1157 | for element in elements: |
||
| 1158 | element.new = False |
||
| 1159 | pickle_file = open('asmxygen.elements.pickle', 'wb') |
||
| 1160 | pickle.dump((elements, id2kind), pickle_file) |
||
| 1161 | pickle_file.close() |
||
| 1162 | |||
| 1163 | if print_stats: |
||
| 1164 | var_count = 0 |
||
| 1165 | mac_count = 0 |
||
| 1166 | lab_count = 0 |
||
| 1167 | fun_count = 0 |
||
| 1168 | uni_count = 0 |
||
| 1169 | str_count = 0 |
||
| 1170 | for element in elements: |
||
| 1171 | if type(element) == AsmVariable: |
||
| 1172 | var_count += 1 |
||
| 1173 | elif type(element) == AsmMacro: |
||
| 1174 | mac_count += 1 |
||
| 1175 | elif type(element) == AsmLabel: |
||
| 1176 | lab_count += 1 |
||
| 1177 | elif type(element) == AsmFunction: |
||
| 1178 | fun_count += 1 |
||
| 1179 | elif type(element) == AsmUnion: |
||
| 1180 | uni_count += 1 |
||
| 1181 | elif type(element) == AsmStruct: |
||
| 1182 | str_count += 1 |
||
| 1183 | print(f'Parsed variable count: {var_count}') |
||
| 1184 | print(f'Parsed macro count: {mac_count}') |
||
| 1185 | print(f'Parsed label count: {lab_count}') |
||
| 1186 | print(f'Parsed function count: {fun_count}') |
||
| 1187 | print(f'Parsed union type count: {uni_count}') |
||
| 1188 | print(f'Parsed structure type count: {str_count}') |
||
| 1189 | |||
| 1190 | if enable_warnings: |
||
| 1191 | open('asmxygen.txt', "w", encoding="utf-8").write(warnings) |
||
| 1192 | |||
| 1193 | if debug_mode: |
||
| 1194 | hash_per_file = "" |
||
| 1195 | for file in output_files: |
||
| 1196 | h = hashlib.sha1(bytes(output_files[file], "ascii")).hexdigest() |
||
| 1197 | hash_per_file += f"{file}: {h}\n" |
||
| 1198 | if not os.path.exists("asmxygen_hash_per_file.txt"): |
||
| 1199 | open("asmxygen_hash_per_file.txt", "w").write(hash_per_file) |
||
| 1200 | print("NEW") |
||
| 1201 | else: |
||
| 1202 | reference_hash_per_file = open("asmxygen_hash_per_file.txt").read() |
||
| 1203 | if reference_hash_per_file != hash_per_file: |
||
| 1204 | diffs = difflib.ndiff(reference_hash_per_file, hash_per_file) |
||
| 1205 | print(''.join(diffs)) |
||
| 1206 | else: |
||
| 1207 | print("SUCCESS") |