WIP: Update Java bindings and add cstest_java implementation

.github/workflows/java-tests.yml

@@ -0,0 +1,87 @@
+name: Java Module CI
+
+on:
+  push:
+  pull_request:
+
+# Stop previous runs on the same branch on new push
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  build:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ubuntu-24.04]
+
+    steps:
+    - uses: actions/checkout@v6
+
+    - uses: actions/setup-java@v5
+      with:
+        distribution: 'temurin'
+        java-version: '21'
+
+    # - name: Setup MSVC
+    #   if: runner.os == 'Windows'
+    #   uses: ilammy/msvc-dev-cmd@v1
+
+    - name: Install JNA
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y libjna-java
+
+    - name: Build and install capstone
+      run: |
+        mkdir build && cd build
+        cmake -DCAPSTONE_INSTALL=1 -DCAPSTONE_BUILD_SHARED_LIBS=1 -DCMAKE_INSTALL_PREFIX=/usr -DCAPSTONE_BUILD_CSTEST=ON ..
+        sudo cmake --build . --config Debug --target install
+
+    - name: Build java bindings
+      working-directory: bindings/java
+      run: make capstone
+
+    - name: Build cstest_java
+      working-directory: bindings/java/cstest_java
+      run: ./gradlew build installDist
+
+    - name: cstest_java integration tests
+      working-directory: suite/cstest/test
+      run: |
+        python3 ./integration_tests.py '../../../bindings/java/cstest_java/app/build/install/app/bin/app'
+
+    - name: cstest_java negative
+      run: |
+        for test_file in \
+          tests/negative/should_fail_I.yaml \
+          tests/negative/should_fail_II.yaml \
+          tests/negative/should_fail_III.yaml \
+          tests/negative/should_fail_IV.yaml \
+          tests/negative/should_fail_V.yaml \
+          tests/negative/should_fail_VI.yaml; do
+          if ./bindings/java/cstest_java/app/build/install/app/bin/app "$test_file"; then
+            echo "Expected failure but test passed: $test_file"
+            exit 1
+          else
+            echo "Correctly failed: $test_file"
+          fi
+        done
+
+    - name: cstest_java MC
+      run: |
+        ./bindings/java/cstest_java/app/build/install/app/bin/app tests/MC/
+
+    - name: cstest_java details
+      run: |
+        ./bindings/java/cstest_java/app/build/install/app/bin/app tests/details/
+
+    - name: cstest_java issues
+      run: |
+        ./bindings/java/cstest_java/app/build/install/app/bin/app tests/issues/
+
+    - name: cstest_java features
+      run: |
+        ./bindings/java/cstest_java/app/build/install/app/bin/app tests/features/

.gitignore

@@ -1,6 +1,7 @@
 files_to_format.txt
 .DS_Store
-
+venv
+.gradle
 # Object files
 *.o
 *.ko

bindings/const_generator.py

@@ -12,12 +12,14 @@
 
 template = {
     'java': {
-            'header': "// For Capstone Engine. AUTO-GENERATED FILE, DO NOT EDIT\npackage capstone;\n\npublic class %s_const {\n",
+            'header': "// For Capstone Engine. AUTO-GENERATED FILE, DO NOT EDIT\npackage capstone;\n\nimport static capstone.Capstone.*;\n\npublic class %s_const {\n",
             'footer': "}",
             'line_format': '\tpublic static final int %s = %s;\n',
+            'line_format_64': '\tpublic static final long %s = %s;\n',
             'out_file': './java/capstone/%s_const.java',
             # prefixes for constant filenames of all archs - case sensitive
             'arm.h': 'Arm',
+            'aarch64.h': ['AArch64', 'AARCH64'],
             'm68k.h': 'M68k',
             'mips.h': 'Mips',
             'x86.h': 'X86',
@@ -29,6 +31,16 @@
             'm680x.h': 'M680x',
             'evm.h': 'Evm',
             'wasm.h': 'Wasm',
+            'mos65xx.h': 'Mos65xx',
+            'bpf.h': 'Bpf',
+            'riscv.h': 'Riscv',
+            'sh.h': 'Sh',
+            'tricore.h': ['Tricore', 'TRICORE'],
+            'alpha.h': ['Alpha', 'ALPHA'],
+            'hppa.h': 'Hppa',
+            'loongarch.h': 'Loongarch',
+            'xtensa.h': 'Xtensa',
+            'arc.h': 'Arc',
             'comment_open': '\t//',
             'comment_close': '',
         },
@@ -134,7 +146,6 @@ def gen(lang):
     global include, INCL_DIR
     print('Generating bindings for', lang)
     templ = template[lang]
-    print('Generating bindings for', lang)
     for target in include:
         if target not in templ:
             print("Warning: No binding found for %s" % target)
@@ -143,7 +154,9 @@ def gen(lang):
         prefixs = []
         if isinstance(prefix, list):
             prefixs = prefix
-            prefix = prefix[0].lower()
+            prefix = prefix[0]
+            if target == 'python':
+                prefix = prefix.lower()
 
         outfile = open(templ['out_file'] %(prefix), 'wb')   # open as binary prevents windows newlines
         outfile.write((templ['header'] % (prefix)).encode("utf-8"))
@@ -311,6 +324,15 @@ def has_special_arch_prefix(x):
                         outfile.write((templ['doc_line_format'] %(doc)).encode("utf-8"))
                 else:
                     line_format = templ['line_format']
+                    if lang == 'java':
+                        try:
+                            value_int = eval(value)
+                            if value_int > 0x7fffffff:
+                                if '1<<' in value:
+                                     value = value.replace('1<<', '1L<<')
+                                line_format = templ['line_format_64']
+                        except Exception:
+                            pass
 
                 outfile.write((line_format %(name, value)).encode("utf-8"))
 

bindings/java/.gitignore

@@ -1,2 +1,2 @@
 *.class
-tags
+*.jar

bindings/java/Makefile

@@ -22,17 +22,25 @@ endif
 
 PYTHON2 ?= python
 
-CAPSTONE_JAVA = Capstone.java Arm_const.java Arm64_const.java Mips_const.java \
+CAPSTONE_JAVA = Capstone.java Arm_const.java AArch64_const.java Mips_const.java \
 				X86_const.java Xcore_const.java Ppc_const.java Sparc_const.java\
-				Sysz_const.java M680x_const.java \
-				Arm.java Arm64.java Mips.java X86.java Xcore.java Ppc.java\
-				Sparc.java Systemz.java M680x.java
+				Systemz_const.java M680x_const.java Riscv_const.java Sh_const.java \
+				Bpf_const.java M68k_const.java TMS320C64x_const.java Evm_const.java \
+				Mos65xx_const.java Wasm_const.java Tricore_const.java Hppa_const.java \
+				Loongarch_const.java Alpha_const.java Xtensa_const.java \
+				Arc_const.java \
+				Arm.java AArch64.java Mips.java X86.java Xcore.java Ppc.java\
+				Sparc.java Systemz.java M680x.java Riscv.java Sh.java Bpf.java \
+				Alpha.java Hppa.java Tricore.java Loongarch.java M68k.java \
+				Evm.java Mos65xx.java Wasm.java TMS320C64x.java Xtensa.java \
+				Arc.java
 
 all: gen_const capstone tests
 
 capstone: capstone_class
 	@mkdir -p $(BLDIR)
 	cd $(OBJDIR) && jar cf $(BLDIR)/capstone.jar capstone/*.class
+	cp $(BLDIR)/capstone.jar cstest_java/app/libs
 
 capstone_class: jna
 ifdef BUILDDIR
@@ -45,7 +53,7 @@ endif
 tests: capstone_class jna
 	@mkdir -p $(OBJDIR)
 	javac -d $(OBJDIR) -classpath "$(JNA):$(BLDIR)/capstone.jar" TestBasic.java\
-		TestArm.java TestArm64.java TestMips.java TestX86.java TestXcore.java\
+		TestArm.java TestAArch64.java TestMips.java TestX86.java TestXcore.java\
 		TestPpc.java TestSparc.java TestSystemz.java TestM680x.java
 
 gen_const:
@@ -62,10 +70,13 @@ ifdef BUILDDIR
 	rm -rf $(OBJDIR)
 endif
 
-TESTS = testbasic arm arm64 m680x mips ppc sparc systemz x86 xcore
+TESTS = testbasic arm aarch64 m680x mips ppc sparc systemz x86 xcore
 check:
 	@for t in $(TESTS); do \
 		echo Check $$t ... ; \
 		./run.sh $$t > /dev/null && echo OK || echo FAILED; \
 	done
 
+cstest: capstone
+	cd cstest_java && ./gradlew run -Djna.library.path=$(shell pwd)/../../build -Djna.debug_load=true --args='--verbosity=info --exclude should_fail_I.yaml --exclude should_fail_II.yaml --exclude should_fail_III.yaml --exclude should_fail_IV.yaml --exclude should_fail_V.yaml --exclude should_fail_VI.yaml'
+

bindings/java/TestArm64.java → bindings/java/TestAArch64.java

@@ -2,11 +2,11 @@
 // By Nguyen Anh Quynh & Dang Hoang Vu,  2013
 
 import capstone.Capstone;
-import capstone.Arm64;
+import capstone.AArch64;
 
-import static capstone.Arm64_const.*;
+import static capstone.AArch64_const.*;
 
-public class TestArm64 {
+public class TestAArch64 {
 
   static byte[] hexString2Byte(String s) {
     // from http://stackoverflow.com/questions/140131/convert-a-string-representation-of-a-hex-dump-to-a-byte-array-using-java
@@ -34,68 +34,66 @@ private static String hex(long i) {
   public static void print_ins_detail(Capstone.CsInsn ins) {
     System.out.printf("0x%x:\t%s\t%s\n", ins.address, ins.mnemonic, ins.opStr);
 
-    Arm64.OpInfo operands = (Arm64.OpInfo) ins.operands;
+    AArch64.OpInfo operands = (AArch64.OpInfo) ins.operands;
 
-    if (operands.op.length != 0) {
-      System.out.printf("\top_count: %d\n", operands.op.length);
-      for (int c=0; c<operands.op.length; c++) {
-        Arm64.Operand i = (Arm64.Operand) operands.op[c];
+    if (operands.operands.length != 0) {
+      System.out.printf("\top_count: %d\n", operands.operands.length);
+      for (int c=0; c<operands.operands.length; c++) {
+        AArch64.Operand i = (AArch64.Operand) operands.operands[c];
         String imm = hex(i.value.imm);
-        if (i.type == ARM64_OP_REG)
+        if (i.type == AARCH64_OP_REG)
           System.out.printf("\t\toperands[%d].type: REG = %s\n", c, ins.regName(i.value.reg));
-        if (i.type == ARM64_OP_REG_MRS)
+        if (i.type == AARCH64_OP_REG_MRS)
           System.out.printf("\t\toperands[%d].type: REG_MRS = 0x%x\n", c, i.value.reg);
-        if (i.type == ARM64_OP_REG_MSR)
+        if (i.type == AARCH64_OP_REG_MSR)
           System.out.printf("\t\toperands[%d].type: REG_MSR = 0x%x\n", c, i.value.reg);
-        if (i.type == ARM64_OP_PSTATE)
-          System.out.printf("\t\toperands[%d].type: PSTATE = 0x%x\n", c, i.value.imm);
-			  if (i.type == ARM64_OP_BARRIER)
-  				System.out.printf("\t\toperands[%d].type: BARRIER = 0x%x\n", c, i.value.imm);
+        // if (i.type == AARCH64_OP_PSTATE)
+        //   System.out.printf("\t\toperands[%d].type: PSTATE = 0x%x\n", c, i.value.imm);
+			  // if (i.type == AARCH64_OP_BARRIER)
+  			// 	System.out.printf("\t\toperands[%d].type: BARRIER = 0x%x\n", c, i.value.imm);
 
-        if (i.type == ARM64_OP_IMM)
+        if (i.type == AARCH64_OP_IMM)
           System.out.printf("\t\toperands[%d].type: IMM = 0x%x\n", c, i.value.imm);
-        if (i.type == ARM64_OP_CIMM)
+        if (i.type == AARCH64_OP_CIMM)
           System.out.printf("\t\toperands[%d].type: C-IMM = %d\n", c, i.value.imm);
-        if (i.type == ARM64_OP_FP)
+        if (i.type == AARCH64_OP_FP)
           System.out.printf("\t\toperands[%d].type: FP = %f\n", c, i.value.fp);
-        if (i.type == ARM64_OP_MEM) {
+        if (i.type == AARCH64_OP_MEM) {
           System.out.printf("\t\toperands[%d].type: MEM\n",c);
-          String base = ins.regName(i.value.mem.base);
-          String index = ins.regName(i.value.mem.index);
-          if (base != null)
-            System.out.printf("\t\t\toperands[%d].mem.base: REG = %s\n", c, base);
-          if (index != null)
-            System.out.printf("\t\t\toperands[%d].mem.index: REG = %s\n", c, index);
+          if (i.value.mem.base != AARCH64_REG_INVALID)
+            System.out.printf("\t\t\toperands[%d].mem.base: REG = %s\n", c, ins.regName(i.value.mem.base));
+          if (i.value.mem.index != AARCH64_REG_INVALID)
+            System.out.printf("\t\t\toperands[%d].mem.index: REG = %s\n", c, ins.regName(i.value.mem.index));
           if (i.value.mem.disp != 0)
             System.out.printf("\t\t\toperands[%d].mem.disp: 0x%x\n", c, i.value.mem.disp);
         }
-        if (i.shift.type != ARM64_SFT_INVALID && i.shift.value > 0)
+        if (i.shift.type != AARCH64_SFT_INVALID && i.shift.value > 0)
           System.out.printf("\t\t\tShift: type = %d, value = %d\n", i.shift.type, i.shift.value);
-        if (i.ext != ARM64_EXT_INVALID)
+        if (i.ext != AARCH64_EXT_INVALID)
           System.out.printf("\t\t\tExt: %d\n", i.ext);
-        if (i.vas != ARM64_VAS_INVALID)
+        if (i.vas != AARCH64LAYOUT_INVALID)
           System.out.printf("\t\t\tVector Arrangement Specifier: 0x%x\n", i.vas);
         if (i.vector_index != -1)
           System.out.printf("\t\t\tVector Index: %d\n", i.vector_index);
 
       }
     }
 
-    if (operands.writeback)
-      System.out.println("\tWrite-back: True");
+    // if (operands.writeback)
+    //   System.out.println("\tWrite-back: True");
 
     if (operands.updateFlags)
       System.out.println("\tUpdate-flags: True");
 
-    if (operands.cc != ARM64_CC_AL && operands.cc != ARM64_CC_INVALID)
+    if (operands.cc != AArch64CC_Invalid)
       System.out.printf("\tCode-condition: %d\n",  operands.cc);
 
   }
 
   public static void main(String argv[]) {
 
     final TestBasic.platform[] all_tests = {
-      new TestBasic.platform(Capstone.CS_ARCH_ARM64, Capstone.CS_MODE_ARM, hexString2Byte(ARM64_CODE), "ARM-64"),
+      new TestBasic.platform(Capstone.CS_ARCH_AARCH64, Capstone.CS_MODE_ARM, hexString2Byte(ARM64_CODE), "ARM-64"),
     };
 
     for (int i=0; i<all_tests.length; i++) {

bindings/java/TestArm.java

@@ -39,10 +39,10 @@ public static void print_ins_detail(Capstone.CsInsn ins) {
 
     Arm.OpInfo operands = (Arm.OpInfo) ins.operands;
 
-    if (operands.op.length != 0) {
-      System.out.printf("\top_count: %d\n", operands.op.length);
-      for (int c=0; c<operands.op.length; c++) {
-        Arm.Operand i = (Arm.Operand) operands.op[c];
+    if (operands.operands.length != 0) {
+      System.out.printf("\top_count: %d\n", operands.operands.length);
+      for (int c=0; c<operands.operands.length; c++) {
+        Arm.Operand i = (Arm.Operand) operands.operands[c];
         String imm = hex(i.value.imm);
         if (i.type == ARM_OP_SYSREG)
           System.out.printf("\t\toperands[%d].type: SYSREG = %d\n", c, i.value.reg);
@@ -70,24 +70,24 @@ public static void print_ins_detail(Capstone.CsInsn ins) {
             System.out.printf("\t\t\toperands[%d].mem.scale: %d\n", c, (i.value.mem.scale));
           if (i.value.mem.disp != 0)
             System.out.printf("\t\t\toperands[%d].mem.disp: 0x%x\n", c, (i.value.mem.disp));
-          if (i.value.mem.lshift != 0)
-            System.out.printf("\t\t\toperands[%d].mem.lshift: 0x%x\n", c, (i.value.mem.lshift));
+          if (i.value.mem.align != 0)
+            System.out.printf("\t\t\toperands[%d].mem.align: 0x%x\n", c, (i.value.mem.align));
         }
         if (i.vector_index > 0)
           System.out.printf("\t\t\toperands[%d].vector_index = %d\n", c, (i.vector_index));
         if (i.shift.type != ARM_SFT_INVALID && i.shift.value > 0)
           System.out.printf("\t\t\tShift: %d = %d\n", i.shift.type, i.shift.value);
-        if (i.subtracted)
+        if (i.subtracted != 0)
           System.out.printf("\t\t\toperands[%d].subtracted = True\n", c);
       }
     }
-    if (operands.writeback)
-      System.out.println("\tWrite-back: True");
+    // if (operands.writeback)
+    //   System.out.println("\tWrite-back: True");
 
     if (operands.updateFlags)
       System.out.println("\tUpdate-flags: True");
 
-    if (operands.cc != ARM_CC_AL && operands.cc != ARM_CC_INVALID)
+    if (operands.cc != ARMCC_AL && operands.cc != ARMCC_UNDEF)
       System.out.printf("\tCode condition: %d\n",  operands.cc);
 
     if (operands.cpsMode > 0)