Add str.center() and str.expandtabs() builtin functions (#2651)

* add center and expandtabs tests

* add _lpython_str_center and _lpython_str_expandtabs

* add _lpython_str_center and _lpython_str_expandtabs

* fix str.center()

* fix typo in line 6924

* update test references

* update tests references

* edit if condition in str.expandtabs

* edit center and expandtabs tests

* update test references

* remove unnecessary print statements

* fix str.expandtabs

* Revert "remove unnecessary print statements"

This reverts commit 1a9a70c.

* edit str.expandtabs

* edit str.expandtabs

* edit str.center

* Formatting changes

* Formatting changes

* Refactor to use len() once

Co-authored-by: Saurabh Kumar <[email protected]>

---------

Co-authored-by: Shaikh Ubaid <[email protected]>
Co-authored-by: Saurabh Kumar <[email protected]>

Author: 3 people

integration_tests/test_str_attributes.py

@@ -472,6 +472,31 @@ def is_numeric():
     assert "".isnumeric() == False
     assert "ab2%3".isnumeric() == False
 
+def center():
+    s: str = "test"
+    assert s.center(8,'*') == "**test**"
+    assert s.center(11) == "    test   "
+    assert s.center(2) == "test"
+    assert s.center(4) == "test"
+    assert s.center(9,'/') == "///test//"
+
+def expandtabs():
+    s: str = '01\t012\t0123\t01234'
+    assert s.expandtabs() == "01      012     0123    01234"
+    assert s.expandtabs(4) == "01  012 0123    01234"
+    assert s.expandtabs(-1) == "01012012301234"
+    s = '\t'
+    assert s.expandtabs() == "        "
+    s = ''
+    assert s.expandtabs() == ""
+    s = '\tThis\ris\na\ttest'
+    assert s.expandtabs(4) == "    This\ris\na   test"
+    s = '\t\t\t'
+    assert s.expandtabs(2) == "      "
+    s = 'test\ttest'
+    assert s.expandtabs(0) == "testtest"
+    assert s.expandtabs(-5) == "testtest"
+
 def check():
     capitalize()
     lower()
@@ -492,6 +517,8 @@ def check():
     is_space()
     is_alnum()
     is_numeric()
+    center()
+    expandtabs()
 
 
 check()

src/lpython/semantics/python_ast_to_asr.cpp

@@ -6880,6 +6880,59 @@ class BodyVisitor : public CommonVisitor<BodyVisitor> {
                 value.m_value = args[2].m_value;
                 fn_args.push_back(al, value);
             }
+        } else if(attr_name == "center") {
+            if (args.size() != 1 && args.size() != 2) {
+                throw SemanticError("str.center() takes one or two argument",
+                    loc);
+            }
+            ASR::expr_t *arg_value = args[0].m_value;
+            ASR::ttype_t *arg_value_type = ASRUtils::expr_type(arg_value);
+            if (!ASRUtils::is_integer(*arg_value_type)) {
+                throw SemanticError("str.center() argument 1 must be integer", loc);
+            }
+
+            fn_call_name = "_lpython_str_center";
+            ASR::call_arg_t str;
+            str.loc = loc;
+            str.m_value = s_var;
+
+            ASR::call_arg_t value;
+            value.loc = loc;
+            value.m_value = args[0].m_value;
+            fn_args.push_back(al, str);
+            fn_args.push_back(al, value);
+
+            if(args.size() == 2){
+                ASR::expr_t *arg_value = args[1].m_value;
+                ASR::ttype_t *arg_value_type = ASRUtils::expr_type(arg_value);
+                if (!ASRUtils::is_character(*arg_value_type)) {
+                    throw SemanticError("str.center() argument 2 must be str", loc);
+                }
+                value.m_value = args[1].m_value;
+                fn_args.push_back(al, value);
+            }
+        } else if(attr_name == "expandtabs") {
+            if(args.size() > 1) {
+                throw SemanticError("str.expandtabs() takes at most one argument.", loc);
+            }
+            fn_call_name = "_lpython_str_expandtabs";
+            ASR::call_arg_t str;
+            str.loc = loc;
+            str.m_value = s_var;
+            fn_args.push_back(al, str);
+
+            if(args.size() == 1){
+                ASR::expr_t *arg_value = args[0].m_value;
+                ASR::ttype_t *arg_value_type = ASRUtils::expr_type(arg_value);
+                if (!ASRUtils::is_integer(*arg_value_type)) {
+                    throw SemanticError("str.expandtabs() argument must be integer", loc);
+                }
+
+                ASR::call_arg_t value;
+                value.loc = loc;
+                value.m_value = args[0].m_value;
+                fn_args.push_back(al, value);
+            }
         } else if(attr_name.size() > 2 && attr_name[0] == 'i' && attr_name[1] == 's') {
             /*
                 String Validation Methods i.e all "is" based functions are handled here

src/lpython/semantics/python_comptime_eval.h

@@ -100,7 +100,9 @@ struct PythonIntrinsicProcedures {
             {"_lpython_str_isupper", {m_builtin, &not_implemented}},
             {"_lpython_str_isdecimal", {m_builtin, &not_implemented}},
             {"_lpython_str_isascii", {m_builtin, &not_implemented}},
-            {"_lpython_str_isspace", {m_builtin, &not_implemented}}
+            {"_lpython_str_isspace", {m_builtin, &not_implemented}},
+            {"_lpython_str_center", {m_builtin, &not_implemented}},
+            {"_lpython_str_expandtabs", {m_builtin, &not_implemented}}
         };
     }
 

src/runtime/lpython_builtin.py

@@ -1085,6 +1085,65 @@ def _lpython_str_isspace(s: str) -> bool:
             return False
     return True
 
+@overload
+def _lpython_str_center(s: str, width: i32, fillchar: str) -> str:
+    """
+    Return centered in a string of length width. 
+    Padding is done using the specified fillchar (default is an ASCII space). 
+    The original string is returned if width is less than or equal to len(s).
+    """
+    if(len(fillchar) != 1):
+        raise TypeError("The fill character must be exactly one character long")
+    str_len: i32 = len(s)
+    if width <= str_len:
+        return s
+    width -= str_len
+    result: str = ""
+    left_padding: i32 = i32(width/2) + _mod(width,2)
+    i: i32 
+    for i in range(left_padding):
+        result += fillchar
+    right_padding: i32 = width - left_padding
+    result += s
+    for i in range(right_padding):
+        result += fillchar
+    return result
+
+@overload
+def _lpython_str_center(s: str, width: i32) -> str:
+    return _lpython_str_center(s, width, ' ')
+
+@overload
+def _lpython_str_expandtabs(s: str, tabsize: i32) -> str:
+    """
+    Return a copy of the string where all tab characters are replaced 
+    by one or more spaces, depending on the current column and the given tab size.
+    """
+    if len(s) == 0:
+        return s
+    col: i32 = 0
+    result: str = ""
+    c: str
+    for c in s:
+        if c == '\t':
+            if tabsize > 0:
+                i: i32
+                iterations: i32 = tabsize - _mod(col,tabsize)
+                for i in range(iterations):
+                    result += ' '
+            col = 0
+        elif c == '\n' or c == '\r':
+            result += c
+            col = 0
+        else:
+            result += c
+            col += 1
+    return result
+
+@overload
+def _lpython_str_expandtabs(s: str) -> str:
+    return _lpython_str_expandtabs(s, 8)
+
 def list(s: str) -> list[str]:
     l: list[str] = []
     i: i32

tests/reference/asr-array_01_decl-39cf894.json

@@ -6,7 +6,7 @@
     "outfile": null,
     "outfile_hash": null,
     "stdout": "asr-array_01_decl-39cf894.stdout",
-    "stdout_hash": "34c5f9983e43e6b5c65f021792e415f0c2e4fe5135c6435eb5322719",
+    "stdout_hash": "292194a8fe4110a90c90bbcbf94f66b70f82978e14108ded75104711",
     "stderr": null,
     "stderr_hash": null,
     "returncode": 0