JPH04184487A - Font cache administration method - Google Patents

Abstract

PURPOSE:To allow the efficient storage of variable size fonts by storing the fonts in the without a wastefull use of memorys font cache region corresponding to the size of the fonts when the fonts are stored in the font cache. CONSTITUTION:A secondary retrieval section 22 retrieves and reads out the fonts meeting character codes from a secondary memory device 14 if the fonts of the prescribed character codes are not stored in the font cache 13. A font side discriminating section 23 discriminates the size of the font read out of the secondary memory device 14 and stores the font into the small-size font cache 13a in the case of the small size below a prescribed size and stores the font into the large-size font cache 13b in the case of the large size above the prescribed size. The variable size fonts are efficiently stored into the memory without a wasteful use of memorys in this way even if the max. size of the variable size fonts are not limited.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology (Figures 14 to 17) Problems to be solved by the invention (Figures 18 to 20) Problems to be solved by the invention Means (Figure 1) Effect (Figure 1) Example/Overall configuration (Figure 2) - Font cache configuration (Figure 3) - Font cache search process (Figure 4) - Individual font cache Search processing (Figures 5 and 6) - Priority processing (Figures 7 and 8) - Font fault processing (Figures 9 to 11) - Secondary search processing (Figures 12 and 13) Effects of the Invention [Summary] The present invention relates to a font cache management method when a computer displays or prints characters on a display device or a printing device using fonts in which each character has a different font size.

The purpose of this method is to provide a font cache management method that can efficiently store variable-size fonts in memory without restricting the maximum size of variable-size fonts. Set up multiple font cache areas,
When storing a font in a font cache, the size of the font is determined and the font is stored in a font cache area according to the size.

[Industrial Field of Application] The present invention relates to a font cache management system, and in particular to a font cache when characters are displayed or printed on a display device or a printing device by a computer using fonts in which each character has a different font size. Regarding management methods.

[Conventional technology] Conventionally, in systems that display or print characters.

It was only possible to display or print characters of a certain size. Such a system has a configuration as shown in FIG. 14, in which a central processing unit (CPU) 1 has a main storage
According to the program stored in the main memory bag, the character code data input from the input device W3 etc. is stored in the main memory bag! 2, a font corresponding to the character code is retrieved from the font data stored in the main storage device 2, and the font is displayed on the display device 4 or printed on the printing device f5.

In this case, the font is usually 24X24 or 16X
This is dot image data consisting of 16 sizes.

An example in the case of 16×16 is shown in FIG. Note that in the following, since a display device and a printing device can be considered the same, only the display device will be explained, but with regard to a printing device, “display” may be replaced with “printing”.

By the way, the font corresponding to the character code is stored in the main memory bag r.
Although it has been stated that the search starts from I12, in reality, it is not possible to store fonts corresponding to all Japanese characters in the main storage device 2. This is because there are over 10,000 characters in Japanese, so even with a 16x16 dot font,
This is because it takes more than 320 Kby8 to store, which takes up space in the main memory.

Furthermore, in order to support multiple fonts such as Mincho and Gothic, a capacity twice or three times this capacity is required, making this capacity problem more pronounced. To this end, a technology called font cache is used to reduce the number of fonts stored in the main storage device 2 at one time.

In font cache management, an area 2a for storing a certain number of fonts is prepared on the main storage device 2, and frequently used fonts are stored in this area.

This area 2a is referred to as a cache or font cache.0 If the font corresponding to the character code to be displayed on the display device 14 is in this cache 2a,
The font is used to read and display characters at high speed. If the font corresponding to the character code does not exist on cache 2a, the secondary storage device! (The font data stored in the at-temperature disk device 6 is retrieved, the font is stored in the cache 2a, and the characters are displayed.)

FIG. 16 and FIG. 17 are explanatory diagrams illustrating such a situation. FIG. 16 shows a case where the font "A" exists in the cache 2a when "A" is displayed on the display device M4. In this case, the font "A" is read from the cache 2a and displayed on the display device M4. ! ! 4 will be displayed. FIG. 17 shows a case where the font [娃J does not exist in the cache 2a when displaying "娃" on the display device f4, and in this case, the font "娃" is displayed on the secondary storage device 6.
is read from and stored in the cache 2a, and
Display device! Note that there is a limit to the number of fonts that can be stored in cache 2a at the same time, so if you want to display a character that is not in the cache when the cache is already full of fonts, display one of the fonts in the cache. and store the new font to be displayed there. In this way, the main memory
Font cache management is a way to manage a large number of characters, such as Japanese, by using the small amount of memory on 2.

[Problems to be Solved by the Invention] Fonts include fixed size fonts and variable size fonts. Variable size fonts are fonts that have become necessary in the world of CAD and desktop publishing. In CAD and desktop publishing, it is required to display characters in the size intended by the user, and for this reason, stroke fonts (vector fonts) or outline fonts that can display characters in any size have been devised. As shown in Figure 18, a stroke font is a font that assumes that characters are composed of line segments, and is a font that consists of the coordinate values of the end points of the line segments, while an outline font considers each character to be a polygon of a figure, and is a font that consists of polygonal fonts. It is a font consisting of coordinate values that form a font, and the more complex the characters, the more information (bytes) is required, and they are collectively called variable-size fonts. These variable-size fonts are made from coordinate values, so the coordinates of line segments and polygons can be converted arbitrarily, so the size of the characters can be adjusted as desired, and it is also easy to display rotated or tilted characters. Therefore, it is a font that is expected to be used more and more in the future.

There is no problem with font cache management when the size of the managed font is fixed, but when the size of the managed font is variable, the following problem occurs.

In other words, if the font is a fixed size, the font shown in Figure 19 (
As shown in a), the cache 2a is divided into units of fixed font size, and fonts can be stored sequentially without waste. However, in the case of variable-size fonts, considering the case where all fonts stored in the cache area are of the maximum size, the cache 2a is divided into units of maximum size as shown in FIG. 19(b), and variable-size fonts are need to be memorized. Therefore, in the case of a fixed font size, there is no wasted space in the cache, but in the case of a variable font size, the maximum size is always required to store one font, and a lot of space is wasted in the cache. There are problems that exist.

FIG. 20 is a size distribution diagram of stroke fonts, which are variable size fonts. As is clear from this distribution diagram, the size of most fonts is less than half of the maximum size. This means that more than half of the area reserved for the main memory bag W2 as the font cache 2a is wasted.

From the above, it is conceivable to suppress the maximum size of a variable size font so that it does not become too large, but in order to do this, there is no choice but to design the font so that the maximum size does not become large.

However, if the size is limited, there is a risk that the characters will not be beautiful, and devices that support user-defined characters will not be able to define the font that the user desires, making it impractical.

From the above, it is an object of the present invention to provide a font cache management method that can efficiently store variable-size fonts in memory without wasting any waste, without limiting the maximum size of variable-size fonts.

[Means for Solving the Problems] Fig. 1 is a diagram explaining the principle of the present invention 6 13 is a font cache, a small size font cache 13a that stores small size fonts,
It has a large size font cache 13b that stores large size fonts. 14 is a secondary storage device such as a magnetic disk device that stores fonts of all characters; 21 is a font cache search unit that searches whether a font corresponding to a given character code is stored in the font cache 13; 22 is a character code 23 is a font size determination unit that determines the size (number of bytes) of the font read from the secondary storage device; Department.

[Operation] The font cache 13 has a plurality of font cache areas 1 corresponding to the font size, for example, a small font cache 13a that stores small fonts of a predetermined size or less, and a large font cache 13a that stores large fonts of a predetermined size or more. A cache 13b is provided.

Then, the variable size font of all characters is stored in secondary storage [1
At the same time, a predetermined number of recently used variable size fonts are stored in the font cache 1 according to the size.
3 is stored in the small size font cache 13a and the large size font cache 13b.

When a predetermined character code occurs in such a state, the font cache search unit 21 searches whether the font of the character code is stored in the font cache 13,
If the font is stored, the font is read from the font cache.

If the font with the predetermined character code is not stored in the font cache 13, the secondary search unit 22 searches and reads out the font corresponding to the character code from the secondary storage [14], and the font size determination unit 23 , determines the size of the font read from the secondary storage device, and stores it in the small size font cache 13a if it is a small size below a predetermined size, and stores it in the large size font cache 13b if it is a large size above a predetermined size. to be memorized.

In this way, when storing multiple font cache areas in the font cache corresponding to the font size and storing fonts in the font cache, the size of the font is determined and the font is stored in the font cache area according to the size. Because you configured it to store variable-size fonts, you don't have to limit the maximum size of variable-size fonts.
To efficiently store variable size fonts in memory without waste.

Furthermore, if fonts are stored in each font cache area in the order in which they were most recently used, it is possible to search in the order of the most likely to be used, thereby improving the efficiency of searching for a desired font.

Furthermore, in each font cache area, a pointer pointing to the next font address, character code, and font coordinate information are stored for each font, as well as an empty font address, the storage address of the most recently used font, and the last one. If configured to store the memory addresses of used fonts, it is possible to easily store the found font by replacing it with the oldest font data, and also to easily store fonts in the font cache area in the order of most recently used fonts. be able to.

[Implementation-Example] For your convenience, Figure 2 is a configuration diagram of a system related to the present invention.
1 is the main control unit [(CPU), 12 is the main storage device, 13
is a font cache reserved in main memory. Font cache 13.

It has a small size font cache 13a that stores fonts smaller than a predetermined size, and a large size font cache 13b that stores fonts larger than a predetermined size. Accordingly, it is stored in the 71% size font cache 13a and the large size font cache 13b.

14 is a secondary storage device such as a magnetic disk device for storing fonts of all characters; 15 is an input device for inputting various data such as character codes; 16 is a display device such as a CRT or liquid crystal display; and 17 is a printing device.

18 is a bus.

The main control unit 1 (CPU) 11 stores character code data input from the input device 13 etc. in the main storage device 12 according to the program stored in the main storage device 12, and stores the font corresponding to the character code in the font cache. 1
3, the corresponding font cache 1
3 and display it on the display device 16 or print it on the printing device 15.

In addition, if the font to be displayed or printed is not stored in the font cache 13, the font stored in the secondary storage device 14 is searched and read out.
5 Determine the size of the read font, and if it is a small size below a predetermined size, store it in the small size font cache 13a, and if it is a large size above a predetermined size, store it in the large size font cache 13b. (This process is called font fault processing.) Thereafter, display or printing is performed using the fonts stored in the font cache 13.

In the above description, two large and small font caches are provided depending on the font size, but the number of font caches is not limited to two, and three or more may be provided depending on the size.

Formation of Font Cache In order to smoothly perform font cache search processing, font fault processing, etc., the present invention uses a small size font cache 13a and a large size font cache 1.
3b is constructed as shown in FIG.

That is, each small size font cache 13 and large size font cache 13
a and 13b are font storage areas 13a-1 and 13b-1 for storing fonts, and empty font address storage areas 13a-2 and 13b for storing empty font addresses, respectively.
13b-2, the highest priority font address storage units 13a-3 and 13b-3 that store the address of the area that stores the most recently used font, and the address of the area that stores the most recently used font. It has the lowest order font addresses 13a-4 and 13b-4.

The font storage areas 13a-1 and 13b-1 are capable of storing a plurality of fonts in the order of latest use, and for each font, a pointer PN pointing to the next font address, a font character code CD, and the like are stored. , font coordinate information FI are stored. If there is no free space, the free font address storage units 13a-2 and 13b-2 contain O.
is stored in the font cache, and if nothing is stored in the font cache, the highest priority font address storage unit 13a-
O is stored in 3, 13b-3, and the pointer (pointer of the oldest used font) in which the area indicated by the lowest order font address is stored is O.

The font storage area 13a-1 of the small-sized font cache 13a is divided into predetermined sizes, and small-sized fonts (pointer PN, font character code CD, and font coordinate information FI) smaller than the specified size are stored in each area. Ru. Further, the font storage area 13b-1 of the large size font cache 13b is divided by maximum font size, and each area stores fonts (pointer PN, font character code CD, and font coordinate information) having a large size larger than a predetermined size. FI) is stored.

Font cache Processing The search process in the entire font cache 13 is performed in the individual font caches 13a and 13b in order, and if a font corresponding to a single character code is found, the search process ends there. do. If a font is not found in one font cache, the next font cache is searched. In this way, if a font is not found even after checking all font caches, font fault processing is performed.

FIG. 4 shows the overall flow of the font cache search process.

When a character code is given and a font of the character code is requested, the number of font caches of different sizes (2 in the example of FIG. 2 since there are two large and small font caches) is set as n (step 101). ).

Then, to determine whether n > O, step 102), n=
If o, the font does not exist in the font cache, so the font fault routine is executed (step 103).

On the other hand, if n Search processing is performed (step 104).

By running individual font cache search routines,
If the font is found in the n-th font cache (step 105), the process ends; if it is not found, count down n (n=n-1).
After that, the process from step 102 onwards is repeated until the corresponding font is found or n=0.

Individual Font Cache Figure 5 is a flowchart of the search process for individual font caches, and Figure 6 is an explanatory diagram thereof.

In each font cache search process, the highest priority font address stored in the storage unit 13a-3 is referred to, and it is checked whether the address is O (step 20).
1) If it is O, it means that the font is not yet stored in the nth font cache.
02), the process returns to step 105 in FIG. 4 and the subsequent processing is performed.

On the other hand, if the highest priority font address is not 0, read the character code CD of the font from the area indicated by the address (step 203), and check whether the character code of the font matches the character code of the character being searched. If they match (step 204), a priority processing routine is executed to give top priority to the search order of the searched font (step 2o5).

In step 204, if there is no match.

Check whether the font address indicated by the pointer is 0. Step 206) - If it is 0, the font being referenced is the last font, and therefore it is determined that the corresponding font does not exist in the nth font cache. (Step 2o7), the process returns to step 105 in FIG. 4 and the subsequent processing is performed. If the font address indicated by the pointer is not 0, step 208 reads the character code of the next font from the area of the font address.
), after which the process from step 204 onward is repeated 6 or more, resulting in the symbol (a) in FIGS. 6(a) and (b).

(b) Or (a), (b), ()1), (d)...
In this order, a check is made to see if the character code of the font matches the search character code.

Figure 7 is a flowchart of priority processing, and Figure 8 is an explanatory diagram of priority processing.

Individual font cache search processing routine (Figure 5)
If a font having a code that matches the search character code is found, it is necessary to give top priority to the search order of the font. That is, it is necessary to update the pointers of each font in the order of the most recently used fonts (priority processing). From the above, if a font having a code matching the search character code is found, the following priority processing is performed. still,
It is assumed that the address of the area where the font whose priority order is changed to the highest priority is stored is A1, the address of the area that stores the pointer pointing to address A1 is AO, and the address pointed to by the pointer of address A1 is A2.

If a font having a code matching the search character code is found, it is determined whether the address A1 of the area where the font is stored is the lowest rank font address (step 301); if it is not the lowest rank font address, then
The highest priority font address stored in the highest priority font address storage unit 13a-3 is saved (step 30
2).

Next, the address A1 indicating the area where the font of the search character code was stored is stored in the highest priority font address storage section 13a-3 (step 303).

After that, the pointer pointing to address A1 (the pointer stored in the area of address AO) is changed to the pointer (A2) stored in the area of address A1 (step 304), and then the pointer of address A1 is changed to the pointer (A2) stored in the area of address A1. The saved value is changed to a certain value (step 305).

FIG. 8 shows how the pointer and the highest priority font address change after priority processing when the font "i" is retrieved.

On the other hand, in step 301 of FIG. 7, address A1 indicating the area where the font of the search character code was stored
If is the lowest priority font address, the highest priority font address stored in the highest priority font address storage section 13a-3 is saved (step 306).

Next, address A1 is stored in the highest priority font address storage section 13a-3 (step 307).

After that, the lowest order font address storage section 13a-4
In addition to storing the address AO of the area for storing the pointer pointing to the address A1 (step 308),
Set the pointer of address AO to 0 (step 309)
.

Next, the pointer at address A1 is changed to a certain value by the above-mentioned save (step 310).

Font fault handling FIG. 9 is a flowchart of font fault processing.

If the font corresponding to the search character code does not exist in the font cache, the font is stored in the secondary storage device 14.
, determine the size of the read font,
If the size is smaller than a predetermined size, the font is stored in the small font cache 13a, and if the font is larger than the predetermined size, the font is stored in the large font cache 13b (font fault processing). That is,
If the font corresponding to the search character code does not exist in the font cache, a secondary search processing routine is executed to check the font corresponding to the character code from the secondary storage device 14 (step 401). In addition, the secondary storage device 14
All fonts are stored in , so you are sure to find the corresponding font.

If a font corresponding to the search character code is found, its size (number of bytes) is determined, it is checked whether the size is larger than a predetermined size, and if it is less than the predetermined size, it is stored in the small size font cache 13a. If the size is larger than a predetermined size, it is stored in the large size font cache 13b (step 402).

- Once the font cache that stores the font is determined, it is checked whether there is space in the font cache (step 403)'. Actually, it checks whether the free font address is ○, and if it is 0, there is no free font address.
If not, it is determined that there is a vacancy.

If there is space, the font is stored in the area pointed to by the free font address, the free address is advanced to the next step (steps 404, 405), and the priority processing routine shown in FIG. 7 is then executed to store the stored font. The search ranking is given top priority, and the pointers of each font are updated in the order of latest use (step 406).

When the priority processing is completed, the font pattern file opened in the secondary storage search processing routine is closed (step 407), and the entire search processing is ended.

FIG. 10 is an explanatory diagram of font fault processing when there is space, and this is a case where the font "i" is newly searched. The free font address now indicates the next free address, and the top priority font address becomes the address of the area where the newly searched font "i" is stored, and the pointer of the newly searched font and "i" is stored. is the address of the area that stores the highest priority font "A" up to that point.

On the other hand, in step 403 of FIG. 9, if there is no free space in the font cache, the font retrieved from the secondary storage device 14 is stored in the area of the oldest retrieved character of the font in the font cache ( Step Step 4o7).

That is, the searched font is stored in the area indicated by the lowest rank font address.

Thereafter, a priority processing routine is executed to give the stored fonts the highest priority in the search order, update the pointers of each font in the order of latest use, and close the font pattern file (steps 406 and 407).
, completes all search processing.

FIG. 11 is an explanatory diagram of font fault processing when there is no space available, and is a case where the font "O" is newly searched. The newly searched font "O" is stored in the area of the oldest searched font "A", the highest priority font address is the address of the area where the font "O" is stored, and the lowest priority address is the font "O". This is the address of the area that stores the font "iJ" that was pointing to "a".

The pointer of the font "i" becomes O, and the pointer of the newly searched font "o" becomes the address of the area storing the previous highest priority font "e".

111 (i) FIG. 12 is a flowchart of the secondary search process, and FIG. 13 is a file configuration diagram in the secondary storage device.

As shown in FIG. 13, the font files in the secondary storage device 14 include a font pattern file FPF and a font index file PIF. The font pattern file FPF is a file that stores font data in code order, and the font index file PIF
is a file that stores the offset (indicating which byte from the beginning of the font pattern file is the font data) of each font data in the font pattern file FPF in code order.

In the secondary search, first open the font index file PIF (step 5o1), then calculate the position where the offset corresponding to the character code is stored based on the size of the character code and offset value (step 502>, character code (2-byte code) is 0~65
535 and the size of the offset value is 4 bytes, the location where the character code offset is stored is determined by the following equation.

Number of bytes from the beginning in the font index file = Character code 503.504)
.

Next, the font index file PIF is closed and the font pattern file FPF is opened in its place (steps 505, 506).

After opening the font pattern file FPF, seek by the offset read in step 504 and read the font data (step 507), return to step 4O2 of font fault processing (Figure 9)9. Although this is a case in which two large size font caches are provided, three or more may be provided by appropriately setting the boundary size.

Although the present invention has been described above with reference to examples, the present invention can be modified in various ways according to the gist of the present invention as described in the claims, and the present invention does not exclude these modifications.

[Effects of the Invention] According to the present invention, when a font cache is provided with a plurality of font cache areas corresponding to the font size and a font is stored in the font cache, the size of the font is determined and the size is adjusted. Since the fonts are stored in corresponding font cache areas, the variable size fonts can be efficiently stored in the memory without any waste, even without restricting the maximum size of the variable size fonts.

Further, according to the present invention, since fonts are stored in each font cache area in the order of most recently used fonts, it is possible to search in the order of the most likely to be used, thereby improving the efficiency of searching for a desired font.

Furthermore, according to the present invention, in each font cache area, a pointer pointing to the next font address, a one-character code, and font coordinate information are stored for each font, as well as an empty font address, the highest priority font address, and the lowest priority font address. Since I configured it to remember the font address,
To easily store a searched font replacing the oldest font data, and to easily store fonts in the font cache area in the order of latest use.

[Brief explanation of the drawing]

Fig. 1 is a diagram of the principle of the present invention, Fig. 2 is a block diagram of a system related to the present invention, Fig. 3 is a block diagram of a font cache, Fig. 4 is a flowchart of the entire font cache search process, and Fig. 5 is a block diagram of a system related to the present invention. Flowchart of individual font cache search processing. FIG. 6 is an explanatory diagram of individual font cache search processing, FIG. 7 is a flowchart of priority processing, and FIG. 8 is an explanatory diagram of priority processing. FIG. 9 is a flowchart of font fault processing. Fig. 10 is an explanatory diagram of fault processing when there is free space, Fig. 11 is an explanatory diagram of fault processing when there is no free space, Fig. 12 is a flowchart of secondary storage search processing, and Fig. 13 is a secondary storage device. Figure 14 is a diagram of the conventional system configuration, Figure 15 is an illustration of font images, Figure 16 is an illustration of font cache management when the relevant font is in the cache, and Figure 17 is applicable to the cache. Figure 18 is an explanatory diagram of ont cache management when there are no fonts, Figure 18 is an explanatory diagram of stroke fonts, and Figure 19 is
The figure is an explanatory diagram of the state of font storage in the cache, and FIG. 20 is a size distribution diagram of stroke fonts. 13...Font cache 13a...Small size font cache 13b...Large size font cache 14...Secondary storage device 21...Font cache search unit 22...Secondary search unit 23...Font size determination unit Tsujun= S = Moon's principle Figure 1 Honest opinion B Month 1: Creating a system Town Hall d Exit Figure 2 (σ) (
bno font~Tsun's mii font A Yashi 1 acceptance of the evening prisoner PU and body> of the tag 4th
Diagram @ Ri's Font ~ Yanyu Aya 1 Elegance, Buried Flowing Water'' η No. 5
Figure Ishimamefuunto ~ Yasshi 1 Fusuma plan
Figure 7 of the general entrance to the power list! (font address
Explanation diagram of the 4 four principles of giving priority to font address Figure 8 Font Follet Ignore A's odd bill F Figure 9 ψ child font end dress
!・Font address depletion Ov If it matches 0 font 1st Retoz Takuri's word ya mouth month η empty Rust b case font foo) Reto@f's proclamation Ming Goku 2 missing right [ri) -
Machi, Se (1 star F ° Evening doctor's Nida L Reen Figure 12 Second Vice-Minister Ko J status fan if) Phi IV Rishi Seiji Figure 13 Condolence rice system system structure ha mouth Figure 14 Font Image Miyata Akimasa Figure 15 Stroke font 0 words ■ Town Figure 18 N Yango Shinhai To font is p) U side 4 / 7 i Tokya 3 ri no kichi tsu 1 Old Figure 16 Iy 7 shi 1 thorns 7 ont -υ) A11 day ~ Nakaunyu'gun day 7 [] Figure 11 (Q)
(b) ~ Font Ryukouji 4 friends W9 and regular call to Yaa A ■
Mouth Figure 19

Claims (3)

[Claims] (1) A variable-size font whose individual size is not fixed is stored in a secondary storage device, and a predetermined number of recently used variable-size fonts are stored in a font cache, and a predetermined character code is generated. , check whether the font of the character code is stored in the font cache, and if it is stored in the font cache, read the font from the font cache,
If the font is not stored, the font cache management method reads the font corresponding to the character code from the secondary storage device and stores it in the font cache. 1. A font cache management method characterized in that, when a font is stored in a font cache, the size of the font is determined and the font is stored in a font cache area according to the size. (2) The font cache management method according to claim 1, wherein fonts are stored in each font cache area in the order in which they were most recently used. (3) Each font cache area stores a pointer pointing to the next font address, character code, and font coordinate information for each font, as well as an area that stores free font addresses and the most recently used font. 2. The font cache management method according to claim 1, wherein the address of the font and the address of an area for storing the last used font are stored.