Merge pull request #68 from zkemail/feat/fix-email-addr-regex

Fixing email address and domain regexes.

packages/apis/src/decomposed_defs/body_hash.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)dkim-signature:"
+      "regex_def": "(\r\n|^)dkim-signature:"
     },
     {
       "is_public": false,

packages/apis/src/decomposed_defs/email_addr.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": true,
-      "regex_def": "[A-Za-z0-9!#$%&'*+=?^_`{|}~.]+@[A-Za-z0-9.-]+"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./@]+@[A-Za-z0-9.\\-]+"
     }
   ]
-}
+}

packages/apis/src/decomposed_defs/email_addr_with_name.json

@@ -6,11 +6,11 @@
     },
     {
       "is_public": true,
-      "regex_def": "[a-zA-Z0-9!#$%&'*+-/=?^_`{\\|}~\\.]+@[a-zA-Z0-9_\\.-]+"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./@]+@[a-zA-Z0-9.\\-]+"
     },
     {
       "is_public": false,
       "regex_def": ">"
     }
   ]
-}
+}

packages/apis/src/decomposed_defs/email_domain.json

@@ -2,15 +2,11 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "[A-Za-z0-9!#$%&'*+-/=?^_`{\\|}~\\.]+"
-    },
-    {
-      "is_public": false,
-      "regex_def": "@"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./]+@"
     },
     {
       "is_public": true,
-      "regex_def": "[A-Za-z0-9\\.-]+"
+      "regex_def": "[A-Za-z0-9.\\-@]+"
     }
   ]
-}
+}

packages/apis/src/decomposed_defs/from_addr.json

@@ -2,19 +2,19 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)from:"
+      "regex_def": "(\r\n|^)from:"
     },
     {
       "is_public": false,
       "regex_def": "([^\r\n]+<)?"
     },
     {
       "is_public": true,
-      "regex_def": "[A-Za-z0-9!#$%&'\\*\\+-/=\\?^_`{\\|}~\\.]+@[A-Za-z0-9\\.-]+"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./@]+@[A-Za-z0-9.\\-]+"
     },
     {
       "is_public": false,
       "regex_def": ">?\r\n"
     }
   ]
-}
+}

packages/apis/src/decomposed_defs/from_all.json

@@ -1,16 +1,16 @@
 {
-  "parts": [
-    {
-      "is_public": false,
-      "regex_def": "((\r\n)|^)from:"
-    },
-    {
-      "is_public": true,
-      "regex_def": "[^\r\n]+"
-    },
-    {
-      "is_public": false,
-      "regex_def": "\r\n"
-    }
-  ]
-}
+    "parts": [
+        {
+            "is_public": false,
+            "regex_def": "(\r\n|^)from:"
+        },
+        {
+            "is_public": true,
+            "regex_def": "[^\r\n]+"
+        },
+        {
+            "is_public": false,
+            "regex_def": "\r\n"
+        }
+    ]
+}

packages/apis/src/decomposed_defs/message_id.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)message-id:"
+      "regex_def": "(\r\n|^)message-id:"
     },
     {
       "is_public": true,

packages/apis/src/decomposed_defs/subject_all.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)subject:"
+      "regex_def": "(\r\n|^)subject:"
     },
     {
       "is_public": true,

packages/apis/src/decomposed_defs/timestamp.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)dkim-signature:"
+      "regex_def": "(\r\n|^)dkim-signature:"
     },
     {
       "is_public": false,

packages/apis/src/decomposed_defs/to_addr.json

@@ -2,19 +2,19 @@
   "parts": [
     {
       "is_public": false,
-      "regex_def": "((\r\n)|^)to:"
+      "regex_def": "(\r\n|^)to:"
     },
     {
       "is_public": false,
       "regex_def": "([^\r\n]+<)?"
     },
     {
       "is_public": true,
-      "regex_def": "[a-zA-Z0-9!#$%&'\\*\\+-/=\\?^_`{\\|}~\\.]+@[a-zA-Z0-9_\\.-]+"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./@]+@[a-zA-Z0-9.\\-]+"
     },
     {
       "is_public": false,
       "regex_def": ">?\r\n"
     }
   ]
-}
+}

packages/apis/src/decomposed_defs/to_all.json

@@ -1,16 +1,16 @@
 {
-  "parts": [
-    {
-      "is_public": false,
-      "regex_def": "((\r\n)|^)to:"
-    },
-    {
-      "is_public": true,
-      "regex_def": "[^\r\n]+"
-    },
-    {
-      "is_public": false,
-      "regex_def": "\r\n"
-    }
-  ]
-}
+    "parts": [
+        {
+            "is_public": false,
+            "regex_def": "(\r\n|^)to:"
+        },
+        {
+            "is_public": true,
+            "regex_def": "[^\r\n]+"
+        },
+        {
+            "is_public": false,
+            "regex_def": "\r\n"
+        }
+    ]
+}

packages/circom/README.md

@@ -2,4 +2,6 @@
 
 Circom circuits for regex verification in [zk-regex](https://github.com/zkemail/zk-regex/tree/main).
 This package contains circom circuits and decomposed regex definitions for common regexes in `./circuits/common` folder.
-
+
+## Note
+Our `email_domain_regex.circom` circuit cannot capture an email address that contains "@" in the name part before the domain part, e.g., "[email protected]@dummy.com", due to limitation of our circuit construction. For example, when "[email protected]@dummy.com" is given, that circuit outputs not "dummy.com" but "[email protected]" as an exposed substring for the domain. However, an adversary cannot exploit this feature to expose a fake domain since the true domain at the end will also be revealed along with it.

packages/circom/circuits/common/email_addr.json

@@ -2,7 +2,7 @@
   "parts": [
     {
       "is_public": true,
-      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\^_`{|}~.]+@[A-Za-z0-9.-]+"
+      "regex_def": "[A-Za-z0-9!#$%&'*+=?\\-\\^_`{|}~./@]+@[A-Za-z0-9.\\-]+"
     }
   ]
-}
+}