Home

Features

Download

Manual

History

Links

History

Version 1.22 Beta:

• Just review my C++ book about dynamic memory allocation. Changed anything that should be allocated on the fly, thus, this version comsume less memory (1160KB comparing 1376KB, not much but not too bad). After quitting, ussualy it will release all the memory, but sometimes it holds 16KB, don't know why, trying to figure out.
• Since I just reduced about 15% of consumed memory, I decided to change the setting for compiling to speed optimization. However, I didn't see any improvement, maybe the change is too small or nothing at all.
• Tried to limit it to a single instance but didn't work on previous version. Just found article 238100 from Microsoft, WinCE use a slightly different methods from desktop version. This version won't run a second instance.
• I was thinking about saving a list of dictionary (like a module list) but too lazy. Thus, extended the MRU (Most Recent Use) list to 8 items and removed the filepath. Treat it as a module list, please!

Version 1.21 Beta:

• Fix the Recent File List, version 1.20B used a file filter for Open File Dialog, thus preventing the files to be added to Recent File List.
• Fix double-tapping on the list, it should set the word in the input box.

Version 1.20 Beta:

• Some characters in Vietnamese are considered alphabetical characters with accent, however, these characters in other languages such as French are considered as whole characters. I added a backup plan for the search engine, just in case there are some exceptions. Thus, if you encounter such a word, that will take a little time to scan through a letter section, but it will be able to find that word in the index.
• Add a file filter for the Open dictionary dialog, it will filter out all the files except dictionary files (*.dz). This idea has been noted at the beginning but I didn't consider it much.

Version 1.19 Beta:

• Some dictionaries ignore the space in their sorting rule, some don't (for example, "a few" would be placed after "able" in some dictionaries). Because of this non-standard thing, I have to change the rough index again. The new version will be able to recognize if space in a dictionary is ignored or not. Sorry guys, you have to spend some little time creating the rough index again.
• Since I store the position of neighboring list, double-tapping on the neighboring list will get the word based on relative index of the word on the list. Thus, no matter what the word is, the search engine won't compare the string, it just returns the definition right at that position. This will reduce the searching time if you double-tapping on the list, amd in worst case, if the search engine for some reason can't read the word, you still can get the definition.
• There is an English-German dictionary which based on the Ding project. This dictionary has some entries at the beginning, started with "- :", "...", and "::" symbol. This kind of sorting was driving me crazy since some dicionaries ignore those symbols. I fixed that one using assumption that other dictionaries will use the same rules. If you find a dictionary that does not following the rule, please let me know so that I can fix it.
• I downloaded a few more dictionaries from www.freedict.de, including English, Africaans, Russian, Swedish, Spanish, Romanian, Italian, Irish, and those just work fine. Since those are small dicionaries, speed is not an issue like Worldnet.
• Current issue: some people having problem running the program. I'm not sure that my BE is different from a typical one or not, so if you are able to run it properly, please send me an email at hoang027@umn.edu. Thanks a lot for your help.

Version 1.18 Beta:

• Change the rough index again. I reallized that after getting the rough index of the index file, I also know how the characters in a language are sorted. This way, the rough index will divide index file into may smaller segments using both the first and second characters. Just try it, you will see a huge improvement in speed (the worst delay time is reduced from 6 seconds to 1 second! ). This took me about 15 hours of coding.
• Add page up and down buttons (with letter P- and P+)
• Add previous and next rough index section buttons, with letter L- and L+, which stand for Letter-/+. My first idea was previous and next letter, but some dictionary just have a huge letter section (such as letter s or t), this will jump too far. So, I decided to let the user see how I divided the index file.

Version 1.17 Beta:

• Make "Look Up" default button, you can tap Enter instead of tapping Look Up button.
• Display SIP when click on input box, hide SIP when input box lose focus or when tap Look Up button.
• Fix a memory error, which created critical error (could hang the BE) on some dictionary with some specific words.

Version 1.16 Beta:

• Fix the font and font size setting. Now the change is applied to the input box, output box, and the neighboring list.
• Add a message in the output box after loading a dictionary to indicate which dictionary has been load, also serve as a success loading indicator.
• The neighboring list should have the word entered in its zoom when displayed.
• Some index files end with a new line character, some end right at the last letter. Just fix the search engine for this variable file ending.

  Note: Delete all the rough index files ( .ridx files) before using the program if you has been using previous version. The program will create new rough index files instead.

Version 1.14 Beta

• Fix a small bug in the search engine (so that it could find one letter unicode word)
• Add the Make Current Dict Default button in Setting dialog. The program will automatically load the default dictionary on start up.
• Neighboring list display 40 instead of 20 words. It also will display the  first 20 words of the dictionary if there is no word has been entered.
• Make the small icon for the program.

Version 1.1 Beta:

• Fix the search engine, (using two rough index arrays instead of one, thus, index-error-proof)
• Add my email to the About box so that I can receive feedback and bug reports (fundamental mistake !!!).

Version 1.0 Beta:

• The first working version since AnyDictz.
• The name is changed to DictzCE. I just want to start from scratch instead of relying on the non-promising AnyDictz design
• The search engine is totally change, using a better aproach with multi-language kept in mind.

AnyDictz 0.11 Beta:

• The very first design of the program, dialog based, since this is my second program using MFC.
• Can only load Worldnet dictionary, nothing is developed yet, even the search engine still has bugs.