only 13 failing encoder tests

This commit is contained in:
Henry Schimke
2022-10-04 18:08:19 -05:00
parent bf0a83c688
commit 86f6bd0c69
5 changed files with 823 additions and 553 deletions

View File

@@ -20,10 +20,11 @@ chrono = "0.4"
chrono-tz = "0.4"
image = {version = "0.24.3", optional = true}
imageproc = {version = "0.23.0", optional = true}
unicode-segmentation = "1.10.0"
[dev-dependencies]
java-properties = "1.4.1"
[features]
default = ["image"]
image = ["dep:image", "dep:imageproc"]
image = ["dep:image", "dep:imageproc"]

View File

@@ -18,6 +18,8 @@ use encoding::EncodingRef;
use lazy_static::lazy_static;
use unicode_segmentation::UnicodeSegmentation;
#[cfg(test)]
mod StringUtilsTestCase;
@@ -2920,7 +2922,7 @@ impl ECIEncoderSet {
* @param fnc1 fnc1 denotes the character in the input that represents the FNC1 character or -1 for a non-GS1 bar
* code. When specified, it is considered an error to pass it as argument to the methods canEncode() or encode().
*/
pub fn new(stringToEncode: &str, priorityCharset: EncodingRef, fnc1: i16) -> Self {
pub fn new(stringToEncodeMain: &str, priorityCharset: EncodingRef, fnc1: Option<&str>) -> Self {
// List of encoders that potentially encode characters not in ISO-8859-1 in one byte.
let mut encoders: Vec<EncodingRef>;
@@ -2928,20 +2930,22 @@ impl ECIEncoderSet {
let mut neededEncoders: Vec<EncodingRef> = Vec::new();
let stringToEncode = stringToEncodeMain.graphemes(true).collect::<Vec<&str>>();
//we always need the ISO-8859-1 encoder. It is the default encoding
neededEncoders.push(encoding::all::ISO_8859_1);
let mut needUnicodeEncoder = priorityCharset.name().starts_with("UTF");
//Walk over the input string and see if all characters can be encoded with the list of encoders
for i in 0..stringToEncode.chars().count() {
for i in 0..stringToEncode.len() {
// for (int i = 0; i < stringToEncode.length(); i++) {
let mut canEncode = false;
for encoder in &neededEncoders {
// for (CharsetEncoder encoder : neededEncoders) {
let c = stringToEncode.chars().nth(i).unwrap();
if c == fnc1 as u8 as char
let c = stringToEncode.get(i).unwrap();
if (fnc1.is_some() && c == fnc1.as_ref().unwrap())
|| encoder
.encode(&c.to_string(), encoding::EncoderTrap::Strict)
.encode(c, encoding::EncoderTrap::Strict)
.is_ok()
{
canEncode = true;
@@ -2950,13 +2954,13 @@ impl ECIEncoderSet {
}
if !canEncode {
//for the character at position i we don't yet have an encoder in the list
for i in 0..ENCODERS.len() {
for i_encoder in 0..ENCODERS.len() {
// for encoder in ENCODERS {
let encoder = ENCODERS.get(i).unwrap();
let encoder = ENCODERS.get(i_encoder).unwrap();
// for (CharsetEncoder encoder : ENCODERS) {
if encoder
.encode(
&stringToEncode.chars().nth(i).unwrap().to_string(),
&stringToEncode.get(i).unwrap(),
encoding::EncoderTrap::Strict,
)
.is_ok()
@@ -3044,15 +3048,15 @@ impl ECIEncoderSet {
return self.priorityEncoderIndex;
}
pub fn canEncode(&self, c: i16, encoderIndex: usize) -> bool {
pub fn canEncode(&self, c: &str, encoderIndex: usize) -> bool {
assert!(encoderIndex < self.len());
let encoder = self.encoders[encoderIndex];
let enc_data = encoder.encode(&c.to_string(), encoding::EncoderTrap::Strict);
let enc_data = encoder.encode(c, encoding::EncoderTrap::Strict);
enc_data.is_ok()
}
pub fn encode_char(&self, c: char, encoderIndex: usize) -> Vec<u8> {
pub fn encode_char(&self, c: &str, encoderIndex: usize) -> Vec<u8> {
assert!(encoderIndex < self.len());
let encoder = self.encoders[encoderIndex];
let enc_data = encoder.encode(&c.to_string(), encoding::EncoderTrap::Strict);
@@ -3101,7 +3105,7 @@ static COST_PER_ECI: usize = 3;
*/
pub struct MinimalECIInput {
bytes: Vec<u16>,
fnc1: i16,
fnc1: u16,
}
impl ECIInput for MinimalECIInput {
@@ -3260,28 +3264,29 @@ impl MinimalECIInput {
* @param fnc1 denotes the character in the input that represents the FNC1 character or -1 if this is not GS1
* input.
*/
pub fn new(stringToEncode: &str, priorityCharset: EncodingRef, fnc1: i16) -> Self {
let encoderSet = ECIEncoderSet::new(stringToEncode, priorityCharset, fnc1);
pub fn new(stringToEncodeInput: &str, priorityCharset: EncodingRef, fnc1: Option<&str>) -> Self {
let stringToEncode = stringToEncodeInput.graphemes(true).collect::<Vec<&str>>();
let encoderSet = ECIEncoderSet::new(stringToEncodeInput, priorityCharset, fnc1);
let bytes = if encoderSet.len() == 1 {
//optimization for the case when all can be encoded without ECI in ISO-8859-1
let mut bytes_hld = vec![0; stringToEncode.len()];
for i in 0..stringToEncode.len() {
// for (int i = 0; i < bytes.length; i++) {
let c = stringToEncode.chars().nth(i).unwrap();
bytes_hld[i] = if c as i16 == fnc1 { 1000 } else { c as u16 };
let c = stringToEncode.get(i).unwrap();
bytes_hld[i] = if fnc1.is_some() && c == fnc1.as_ref().unwrap() { 1000 } else { c.chars().nth(0).unwrap() as u16 };
}
bytes_hld
} else {
Self::encodeMinimally(stringToEncode, &encoderSet, fnc1)
Self::encodeMinimally(stringToEncodeInput, &encoderSet, fnc1.as_ref().unwrap().chars().nth(0).unwrap() as u16)
};
Self {
bytes: bytes,
fnc1: fnc1,
fnc1: fnc1.as_ref().unwrap().chars().nth(0).unwrap() as u16,
}
}
pub fn getFNC1Character(&self) -> i16 {
pub fn getFNC1Character(&self) -> u16 {
self.fnc1
}
@@ -3321,14 +3326,15 @@ impl MinimalECIInput {
edges: &mut Vec<Vec<Option<Rc<InputEdge>>>>,
from: usize,
previous: Option<Rc<InputEdge>>,
fnc1: i16,
fnc1: u16,
) {
let ch = stringToEncode.chars().nth(from).unwrap() as i16;
// let ch = stringToEncode.chars().nth(from).unwrap() as i16;
let ch = stringToEncode.graphemes(true).nth(from).unwrap();
let mut start = 0;
let mut end = encoderSet.len();
if encoderSet.getPriorityEncoderIndex() >= 0
&& (ch as i16 == fnc1 || encoderSet.canEncode(ch, encoderSet.getPriorityEncoderIndex()))
&& (ch.chars().nth(0).unwrap() as u16 == fnc1 || encoderSet.canEncode(ch, encoderSet.getPriorityEncoderIndex()))
{
start = encoderSet.getPriorityEncoderIndex();
end = start + 1;
@@ -3336,7 +3342,7 @@ impl MinimalECIInput {
for i in start..end {
// for (int i = start; i < end; i++) {
if ch as i16 == fnc1 || encoderSet.canEncode(ch, i) {
if ch.chars().nth(0).unwrap() as u16 == fnc1 || encoderSet.canEncode(ch, i) {
Self::addEdge(
edges,
from + 1,
@@ -3349,7 +3355,7 @@ impl MinimalECIInput {
pub fn encodeMinimally(
stringToEncode: &str,
encoderSet: &ECIEncoderSet,
fnc1: i16,
fnc1: u16,
) -> Vec<u16> {
let inputLength = stringToEncode.len();
@@ -3395,7 +3401,7 @@ impl MinimalECIInput {
intsAL.splice(0..0, [1000]);
} else {
let bytes: Vec<u16> = encoderSet
.encode_char(c.c as u8 as char, c.encoderIndex)
.encode_char(&c.c , c.encoderIndex)
.iter()
.map(|x| *x as u16)
.collect();
@@ -3429,25 +3435,27 @@ impl MinimalECIInput {
}
struct InputEdge {
c: i16,
c: String,
encoderIndex: usize, //the encoding of this edge
previous: Option<Rc<InputEdge>>,
cachedTotalSize: usize,
}
impl InputEdge {
pub fn new(
c: i16,
c: &str,
encoderSet: &ECIEncoderSet,
encoderIndex: usize,
previous: Option<Rc<InputEdge>>,
fnc1: i16,
fnc1: u16,
) -> Self {
let mut size = if c == 1000 {
let mut size = if c == "\u{1000}" {
1
} else {
encoderSet.encode_char(c as u8 as char, encoderIndex).len()
encoderSet.encode_char(c, encoderIndex).len()
};
let fnc1Str = String::from_utf16(&[fnc1]).unwrap();
if let Some(prev) = previous {
let previousEncoderIndex = prev.encoderIndex;
if previousEncoderIndex != encoderIndex {
@@ -3456,7 +3464,7 @@ impl InputEdge {
size += prev.cachedTotalSize;
Self {
c: if c as i16 == fnc1 { 1000 } else { c as i16 },
c: if c == fnc1Str { String::from("\u{1000}") } else { String::from(c) },
encoderIndex,
previous: Some(prev.clone()),
cachedTotalSize: size,
@@ -3468,7 +3476,7 @@ impl InputEdge {
}
Self {
c: if c as i16 == fnc1 { 1000 } else { c as i16 },
c: if c == fnc1Str { String::from("\u{1000}") } else { String::from(c) },
encoderIndex,
previous: None,
cachedTotalSize: size,
@@ -3502,7 +3510,7 @@ impl InputEdge {
}
pub fn isFNC1(&self) -> bool {
self.c == 1000
self.c == "\u{1000}"
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -357,9 +357,12 @@ fn chooseModeWithEncoding(content: &str, encoding: EncodingRef) -> Mode {
}
pub fn isOnlyDoubleByteKanji(content: &str) -> bool {
let bytes = SHIFT_JIS_CHARSET
.encode(content, encoding::EncoderTrap::Strict)
.expect("encode");
let bytes = if let Ok(byt) = SHIFT_JIS_CHARSET
.encode(content, encoding::EncoderTrap::Strict) {
byt
}else {
return false
};
let length = bytes.len();
if length % 2 != 0 {
return false;
@@ -367,7 +370,7 @@ pub fn isOnlyDoubleByteKanji(content: &str) -> bool {
let mut i = 0;
while i < length {
// for (int i = 0; i < length; i += 2) {
let byte1 = bytes[i] & 0xFF;
let byte1 = bytes[i];
if (byte1 < 0x81 || byte1 > 0x9F) && (byte1 < 0xE0 || byte1 > 0xEB) {
return false;
}

View File

@@ -24,6 +24,8 @@ use crate::{
Exceptions,
};
use unicode_segmentation::{Graphemes, UnicodeSegmentation};
use super::encoder;
pub enum VersionSize {
@@ -84,7 +86,7 @@ impl fmt::Display for VersionSize {
* @author Alex Geller
*/
pub struct MinimalEncoder {
stringToEncode: String,
stringToEncode: Vec<String>,
isGS1: bool,
encoders: ECIEncoderSet,
ecLevel: ErrorCorrectionLevel,
@@ -110,9 +112,12 @@ impl MinimalEncoder {
ecLevel: ErrorCorrectionLevel,
) -> Self {
Self {
stringToEncode: String::from(stringToEncode),
stringToEncode: stringToEncode
.graphemes(true)
.map(|p| p.to_owned())
.collect::<Vec<String>>(),
isGS1,
encoders: ECIEncoderSet::new(&stringToEncode, priorityCharset, -1),
encoders: ECIEncoderSet::new(&stringToEncode, priorityCharset, None),
ecLevel,
}
@@ -222,23 +227,34 @@ impl MinimalEncoder {
// }
}
pub fn isNumeric(c: char) -> bool {
return c >= '0' && c <= '9';
pub fn isNumeric(c: &str) -> bool {
if c.len() == 1 {
let ch = c.chars().nth(0).unwrap();
ch >= '0' && ch <= '9'
} else {
false
}
// return c >= '0' && c <= '9';
}
pub fn isDoubleByteKanji(c: char) -> bool {
return encoder::isOnlyDoubleByteKanji(&String::from(c));
pub fn isDoubleByteKanji(c: &str) -> bool {
return encoder::isOnlyDoubleByteKanji(&c);
}
pub fn isAlphanumeric(c: char) -> bool {
return encoder::getAlphanumericCode(c as u8 as u32) != -1;
pub fn isAlphanumeric(c: &str) -> bool {
if c.len() == 1 {
let ch = c.chars().nth(0).unwrap();
encoder::getAlphanumericCode(ch as u32) != -1
} else {
false
}
// return encoder::getAlphanumericCode(c as u8 as u32) != -1;
}
pub fn canEncode(&self, mode: &Mode, c: char) -> bool {
pub fn canEncode(&self, mode: &Mode, c: &str) -> bool {
match mode {
Mode::NUMERIC => Self::isNumeric(c),
Mode::ALPHANUMERIC => Self::isAlphanumeric(c),
Mode::STRUCTURED_APPEND => todo!(),
Mode::BYTE => true,
Mode::KANJI => Self::isDoubleByteKanji(c),
_ => false, // any character can be encoded as byte(s). Up to the caller to manage splitting into
@@ -305,7 +321,8 @@ impl MinimalEncoder {
let priorityEncoderIndex = self.encoders.getPriorityEncoderIndex();
if priorityEncoderIndex >= 0
&& self.encoders.canEncode(
self.stringToEncode.chars().nth(from as usize).unwrap() as i16,
// self.stringToEncode.chars().nth(from as usize).unwrap() as i16,
&self.stringToEncode[from as usize],
priorityEncoderIndex,
)
{
@@ -317,7 +334,7 @@ impl MinimalEncoder {
// for (int i = start; i < end; i++) {
if self
.encoders
.canEncode(self.stringToEncode.chars().nth(from).unwrap() as i16, i)
.canEncode(&self.stringToEncode.get(from).unwrap(), i)
{
self.addEdge(
edges,
@@ -330,13 +347,13 @@ impl MinimalEncoder {
previous.clone(),
version,
self.encoders.clone(),
&self.stringToEncode,
self.stringToEncode.clone(),
))),
);
}
}
if self.canEncode(&Mode::KANJI, self.stringToEncode.chars().nth(from).unwrap()) {
if self.canEncode(&Mode::KANJI, &self.stringToEncode.get(from).unwrap()) {
self.addEdge(
edges,
from,
@@ -348,16 +365,13 @@ impl MinimalEncoder {
previous.clone(),
version,
self.encoders.clone(),
&self.stringToEncode,
self.stringToEncode.clone(),
))),
);
}
let inputLength = self.stringToEncode.len();
if self.canEncode(
&Mode::ALPHANUMERIC,
self.stringToEncode.chars().nth(from).unwrap(),
) {
if self.canEncode(&Mode::ALPHANUMERIC, self.stringToEncode.get(from).unwrap()) {
self.addEdge(
edges,
from,
@@ -368,7 +382,7 @@ impl MinimalEncoder {
if from + 1 >= inputLength
|| !self.canEncode(
&Mode::ALPHANUMERIC,
self.stringToEncode.chars().nth(from + 1).unwrap(),
self.stringToEncode.get(from + 1).unwrap(),
)
{
1
@@ -378,15 +392,12 @@ impl MinimalEncoder {
previous.clone(),
version,
self.encoders.clone(),
&self.stringToEncode,
self.stringToEncode.clone(),
))),
);
}
if self.canEncode(
&Mode::NUMERIC,
self.stringToEncode.chars().nth(from).unwrap(),
) {
if self.canEncode(&Mode::NUMERIC, self.stringToEncode.get(from).unwrap()) {
self.addEdge(
edges,
from,
@@ -395,17 +406,15 @@ impl MinimalEncoder {
from,
0,
if from + 1 >= inputLength
|| !self.canEncode(
&Mode::NUMERIC,
self.stringToEncode.chars().nth(from + 1).unwrap(),
)
|| !self
.canEncode(&Mode::NUMERIC, self.stringToEncode.get(from + 1).unwrap())
{
1
} else {
if from + 2 >= inputLength
|| !self.canEncode(
&Mode::NUMERIC,
self.stringToEncode.chars().nth(from + 2).unwrap(),
self.stringToEncode.get(from + 2).unwrap(),
)
{
2
@@ -416,7 +425,7 @@ impl MinimalEncoder {
previous.clone(),
version,
self.encoders.clone(),
&self.stringToEncode,
self.stringToEncode.clone(),
))),
);
}
@@ -535,6 +544,7 @@ impl MinimalEncoder {
* The encodation ECI(UTF-8),BYTE(XXYY) is longer with a size of 88.
*/
// let inputLength = self.stringToEncode.chars().count();
let inputLength = self.stringToEncode.len();
// Array that represents vertices. There is a vertex for every character, encoding and mode. The vertex contains
@@ -580,6 +590,9 @@ impl MinimalEncoder {
return Err(Exceptions::WriterException(format!(
r#"Internal error: failed to encode "{}"#,
self.stringToEncode
.iter()
.map(|x| String::from(x))
.collect::<String>() //fold("", |acc,x| [acc,&x].concat())
)));
}
Ok(RXingResultList::new(
@@ -591,7 +604,7 @@ impl MinimalEncoder {
self.isGS1,
&self.ecLevel,
self.encoders.clone(),
&self.stringToEncode,
self.stringToEncode.clone(),
))
}
}
@@ -604,7 +617,7 @@ pub struct Edge {
previous: Option<Rc<Edge>>,
cachedTotalSize: u32,
encoders: ECIEncoderSet,
stringToEncode: String,
stringToEncode: Vec<String>,
}
impl Edge {
pub fn new(
@@ -615,7 +628,7 @@ impl Edge {
previous: Option<Rc<Edge>>,
version: VersionRef,
encoders: ECIEncoderSet,
stringToEncode: &'_ str,
stringToEncode: Vec<String>,
) -> Self {
let nci = if mode == Mode::BYTE || previous.is_none() {
charsetEncoderIndex
@@ -628,7 +641,7 @@ impl Edge {
charsetEncoderIndex: nci,
characterLength,
previous: previous.clone(),
stringToEncode: String::from(stringToEncode),
stringToEncode: stringToEncode.clone(),
cachedTotalSize: {
let mut size = if previous.is_some() {
previous.as_ref().unwrap().cachedTotalSize
@@ -657,13 +670,22 @@ impl Edge {
}
Mode::ALPHANUMERIC => size += if characterLength == 1 { 6 } else { 11 },
Mode::BYTE => {
let n: String = stringToEncode
.iter()
.skip(fromPosition as usize)
.take(characterLength as usize)
.map(|x| String::from(x))
.collect();
size += 8 * encoders
.encode_string(
&stringToEncode[fromPosition as usize
..(fromPosition + characterLength as usize)],
charsetEncoderIndex as usize,
)
.encode_string(&n, charsetEncoderIndex as usize)
.len() as u32;
// size += 8 * encoders
// .encode_string(
// &stringToEncode[fromPosition as usize
// ..(fromPosition + characterLength as usize)],
// charsetEncoderIndex as usize,
// )
// .len() as u32;
if needECI {
size += 4 + 8; // the ECI assignment numbers for ISO-8859-x, UTF-8 and UTF-16 are all 8 bit long
}
@@ -743,7 +765,7 @@ impl RXingResultList {
isGS1: bool,
ecLevel: &ErrorCorrectionLevel,
encoders: ECIEncoderSet,
stringToEncode: &str,
stringToEncode: Vec<String>,
) -> Self {
let mut length = 0;
let mut current = Some(solution);
@@ -772,7 +794,7 @@ impl RXingResultList {
current.as_ref().unwrap().charsetEncoderIndex,
length,
encoders.clone(),
stringToEncode,
stringToEncode.clone(),
version,
));
length = 0;
@@ -785,7 +807,7 @@ impl RXingResultList {
current.as_ref().unwrap().charsetEncoderIndex,
0,
encoders.clone(),
stringToEncode,
stringToEncode.clone(),
version,
));
}
@@ -805,23 +827,41 @@ impl RXingResultList {
0,
0,
encoders.clone(),
stringToEncode,
stringToEncode.clone(),
version,
));
}
}
let first = list.get(0);
// prepend or insert a FNC1_FIRST_POSITION after the ECI (if any)
if first.is_some() && first.as_ref().unwrap().mode != Mode::ECI && containsECI {
list.push(RXingResultNode::new(
Mode::FNC1_FIRST_POSITION,
0,
0,
0,
encoders.clone(),
stringToEncode,
version,
));
if first.is_some() && first.as_ref().unwrap().mode != Mode::ECI {//&& containsECI {
list.insert(
if first.as_ref().unwrap().mode != Mode::ECI {
//first
list.len()
} else {
//second
list.len()-1
},
RXingResultNode::new(
Mode::FNC1_FIRST_POSITION,
0,
0,
0,
encoders.clone(),
stringToEncode.clone(),
version,
),
);
// list.push(RXingResultNode::new(
// Mode::FNC1_FIRST_POSITION,
// 0,
// 0,
// 0,
// encoders.clone(),
// stringToEncode.clone(),
// version,
// ));
}
}
@@ -937,7 +977,7 @@ struct RXingResultNode {
characterLength: u32,
encoders: ECIEncoderSet,
version: VersionRef,
stringToEncode: String,
stringToEncode: Vec<String>,
}
impl RXingResultNode {
@@ -947,7 +987,7 @@ impl RXingResultNode {
charsetEncoderIndex: usize,
characterLength: u32,
encoders: ECIEncoderSet,
stringToEncode: &str,
stringToEncode: Vec<String>,
version: VersionRef,
) -> Self {
Self {
@@ -956,7 +996,7 @@ impl RXingResultNode {
charsetEncoderIndex,
characterLength,
encoders,
stringToEncode: String::from(stringToEncode),
stringToEncode,
version,
}
}
@@ -1023,8 +1063,9 @@ impl RXingResultNode {
if self.mode == Mode::BYTE {
self.encoders
.encode_string(
&self.stringToEncode[self.fromPosition as usize
..(self.fromPosition + self.characterLength as usize)],
// &self.stringToEncode[self.fromPosition as usize
// ..(self.fromPosition + self.characterLength as usize)],
&self.stringToEncode.get(self.fromPosition as usize).unwrap(),
self.charsetEncoderIndex as usize,
)
.len() as u32
@@ -1053,8 +1094,9 @@ impl RXingResultNode {
} else if self.characterLength > 0 {
// append data
encoder::appendBytes(
&self.stringToEncode[self.fromPosition as usize
..(self.fromPosition + self.characterLength as usize)],
// &self.stringToEncode[self.fromPosition as usize
// ..(self.fromPosition + self.characterLength as usize)],
&self.stringToEncode.get(self.fromPosition).unwrap(),
self.mode,
bits,
self.encoders.getCharset(self.charsetEncoderIndex as usize),
@@ -1067,7 +1109,7 @@ impl RXingResultNode {
let mut result = String::new();
for i in 0..s.chars().count() {
// for (int i = 0; i < s.length(); i++) {
if (s.chars().nth(i).unwrap() as u8) < 32 || (s.chars().nth(i).unwrap() as u8) > 126 {
if (s.chars().nth(i).unwrap() as u32) < 32 || (s.chars().nth(i).unwrap() as u32) > 126 {
result.push('.');
} else {
result.push(s.chars().nth(i).unwrap());
@@ -1089,10 +1131,18 @@ impl fmt::Display for RXingResultNode {
.name(),
);
} else {
result.push_str(&Self::makePrintable(
&self.stringToEncode[self.fromPosition as usize
..(self.fromPosition + self.characterLength as usize)],
));
let sub_string: String = self
.stringToEncode
.iter()
.skip(self.fromPosition as usize)
.take(self.characterLength as usize)
.map(|x| String::from(x))
.collect();
// result.push_str(&Self::makePrintable(
// &self.stringToEncode[self.fromPosition as usize
// ..(self.fromPosition + self.characterLength as usize)],
// ));
result.push_str(&Self::makePrintable(&sub_string));
}
result.push(')');