Fix several issues in isHostAndPort, isHostname, and isUri

src/main/java/build/buf/protovalidate/CustomOverload.java

@@ -16,7 +16,6 @@
 
 import com.google.common.primitives.Bytes;
 import java.util.HashSet;
-import java.util.Locale;
 import java.util.Set;
 import java.util.regex.Pattern;
 import org.projectnessie.cel.common.types.BoolT;
@@ -463,6 +462,10 @@ private static boolean isPort(String str) {
       return false;
     }
 
+    if (str.length() > 1 && str.charAt(0) == '0') {
+      return false;
+    }
+
     for (int i = 0; i < str.length(); i++) {
       char c = str.charAt(i);
       if ('0' <= c && c <= '9') {
@@ -557,7 +560,7 @@ private static boolean isHostname(String val) {
 
     boolean allDigits = false;
 
-    String[] parts = str.toLowerCase(Locale.getDefault()).split("\\.", -1);
+    String[] parts = str.split("\\.", -1);
 
     // split hostname on '.' and validate each part
     for (String part : parts) {
@@ -572,8 +575,8 @@ private static boolean isHostname(String val) {
       // for each character in part
       for (int i = 0; i < part.length(); i++) {
         char c = part.charAt(i);
-        // if the character is not a-z, 0-9, or '-', it is invalid
-        if ((c < 'a' || c > 'z') && (c < '0' || c > '9') && c != '-') {
+        // if the character is not a-z, A-Z, 0-9, or '-', it is invalid
+        if ((c < 'a' || c > 'z') && (c < 'A' || c > 'Z') && (c < '0' || c > '9') && c != '-') {
           return false;
         }
 

src/main/java/build/buf/protovalidate/Uri.java

@@ -167,7 +167,7 @@ private boolean scheme() {
     if (this.alpha()) {
       while (this.alpha() || this.digit() || this.take('+') || this.take('-') || this.take('.')) {}
 
-      if (this.str.charAt(this.index) == ':') {
+      if (this.peek(':')) {
         return true;
       }
     }
@@ -253,10 +253,8 @@ private boolean userinfo() {
         continue;
       }
 
-      if (this.index < this.str.length()) {
-        if (this.str.charAt(this.index) == '@') {
-          return true;
-        }
+      if (this.peek('@')) {
+        return true;
       }
 
       this.index = start;
@@ -335,15 +333,11 @@ private boolean checkHostPctEncoded(String str) {
    * <pre>host = IP-literal / IPv4address / reg-name.
    */
   private boolean host() {
-    if (this.index >= this.str.length()) {
-      return true;
-    }
-
     int start = this.index;
     this.pctEncodedFound = false;
 
     // Note: IPv4address is a subset of reg-name
-    if ((this.str.charAt(this.index) == '[' && this.ipLiteral()) || this.regName()) {
+    if ((this.peek('[') && this.ipLiteral()) || this.regName()) {
       if (this.pctEncodedFound) {
         String rawHost = this.str.substring(start, this.index);
         // RFC 3986:
@@ -446,7 +440,7 @@ private boolean ipv6Address() {
   /**
    * Determines whether the current position is a valid IPv6addrz.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>IPv6addrz = IPv6address "%25" ZoneID
    */
@@ -465,7 +459,7 @@ private boolean ipv6Addrz() {
   /**
    * Determines whether the current position is a valid zone ID.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>ZoneID = 1*( unreserved / pct-encoded )
    */
@@ -486,7 +480,7 @@ private boolean zoneID() {
   /**
    * Determines whether the current position is a valid IPvFuture.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>IPvFuture  = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" )
    */
@@ -517,7 +511,7 @@ private boolean ipvFuture() {
   /**
    * Determines whether the current position is a valid reg-name.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>reg-name = *( unreserved / pct-encoded / sub-delims )
    *
@@ -536,7 +530,7 @@ private boolean regName() {
         return true;
       }
 
-      if (this.str.charAt(this.index) == ':') {
+      if (this.peek(':')) {
         return true;
       }
 
@@ -570,7 +564,7 @@ private boolean isPathEnd() {
   /**
    * Determines whether the current position is a valid path-abempty.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>path-abempty = *( "/" segment )
    *
@@ -593,7 +587,7 @@ private boolean pathAbempty() {
   /**
    * Determines whether the current position is a valid path-absolute.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>path-absolute = "/" [ segment-nz *( "/" segment ) ]
    *
@@ -620,7 +614,7 @@ private boolean pathAbsolute() {
   /**
    * Determines whether the current position is a valid path-noscheme.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>path-noscheme = segment-nz-nc *( "/" segment )
    *
@@ -645,7 +639,7 @@ private boolean pathNoscheme() {
   /**
    * Determines whether the current position is a valid path-rootless.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>path-rootless = segment-nz *( "/" segment )
    *
@@ -670,7 +664,7 @@ private boolean pathRootless() {
   /**
    * Determines whether the current position is a valid path-empty.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>path-empty = 0<pchar>
    *
@@ -683,7 +677,7 @@ private boolean pathEmpty() {
   /**
    * Determines whether the current position is a valid segment.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>segment = *pchar
    */
@@ -696,7 +690,7 @@ private boolean segment() {
   /**
    * Determines whether the current position is a valid segment-nz.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>segment-nz = 1*pchar
    */
@@ -716,7 +710,7 @@ private boolean segmentNz() {
   /**
    * Determines whether the current position is a valid segment-nz-nc.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" )
    *                   ; non-zero-length segment without any colon ":"
@@ -738,7 +732,7 @@ private boolean segmentNzNc() {
   /**
    * Determines whether the current position is a valid pchar.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>pchar = unreserved / pct-encoded / sub-delims / ":" / "@"
    */
@@ -753,7 +747,7 @@ private boolean pchar() {
   /**
    * Determines whether the current position is a valid query.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>query = *( pchar / "/" / "?" )
    *
@@ -767,7 +761,7 @@ private boolean query() {
         continue;
       }
 
-      if (this.index == this.str.length() || this.str.charAt(this.index) == '#') {
+      if (this.peek('#') || this.index == this.str.length()) {
         return true;
       }
 
@@ -780,7 +774,7 @@ private boolean query() {
   /**
    * Determines whether the current position is a valid fragment.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>fragment = *( pchar / "/" / "?" )
    *
@@ -807,7 +801,7 @@ private boolean fragment() {
   /**
    * Determines whether the current position is a valid pct-encoded.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>pct-encoded = "%"+HEXDIG+HEXDIG
    *
@@ -830,7 +824,7 @@ private boolean pctEncoded() {
   /**
    * Determines whether the current position is an unreserved character.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~"
    */
@@ -846,7 +840,7 @@ private boolean unreserved() {
   /**
    * Determines whether the current position is a sub-delim.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>sub-delims  = "!" / "$" / "&" / "'" / "(" / ")"
    *                  / "*" / "+" / "," / ";" / "="
@@ -868,7 +862,7 @@ private boolean subDelims() {
   /**
    * Determines whether the current position is an alpha character.
    *
-   * Parses the rule:
+   * <p>Parses the rule:
    *
    * <pre>ALPHA =  %x41-5A / %x61-7A ; A-Z / a-z
    */
@@ -942,4 +936,8 @@ private boolean take(char c) {
 
     return false;
   }
+
+  private boolean peek(char c) {
+    return this.index < this.str.length() && this.str.charAt(this.index) == c;
+  }
 }