JP3411350B2 - Outline character expansion device - Google Patents

Description

Detailed Description of the Invention

(Table of contents) Industrial applications Conventional technology (FIGS. 24 to 26) Problems to be Solved by the Invention (FIGS. 27 to 30) Means for Solving the Problems (FIGS. 1 to 3) Action Example (A) Description of system to which the present invention is applied (FIG. 4) (B) Description of the first embodiment (FIGS. 5 to 11) (C) Description of the second embodiment (FIGS. 12 to 20) (D) Description of the third embodiment (FIGS. 21 to 23) The invention's effect

[0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for speeding up raster character expansion of outline characters in a printing device or a display device.

[0003]

2. Description of the Related Art Generally, the size of a character printed by a printing device is predetermined such as 7, 9 or 12 points, and a character pattern is registered for each size. In this case, it is not possible to print characters of unregistered size. Therefore, in order to print characters of various sizes, it is necessary to register a character pattern for each size, and the amount of registered data in that case becomes enormous.

Therefore, conventionally, a character pattern obtained by rasterizing an outline character into a desired character size is printed. In outline characters, each vertex coordinate on the outline (outline) of the character is stored as character data,
By arranging these coordinates according to the desired character size [scaling (scaling) processing], connecting the vertices [outline drawing processing], and filling the inside thereof [surface painting processing],
A raster character (character pattern) having a desired character size can be obtained. Such processing is called raster character expansion of outline characters.

By the way, if such an outline character is expanded every time each character pattern is printed,
Since this process requires time, a pipeline method that performs the above-mentioned various processes in parallel is adopted, and once expanded character patterns are retained in the cache according to their letter addresses (pattern addresses). By doing so, the character pattern that has already been raster-expanded can be reused, and the speed of character processing is increased.

Therefore, normally, as shown in FIGS. 25 to 28, two sets (primary and secondary) of the character expansion cache tables 21A and 21B and the cache management tables 22A and 22B are provided in the character pattern cache control unit 20. ing. Here, the character expansion cache tables 21A and 21B store character patterns obtained by expanding outline characters from character data according to a predetermined character size.

Further, the cache management tables 22A, 2
2B is a character expansion cache table 21
A character address and character size for each character pattern stored in the character expansion cache tables 21A and 21B and expansion destination addresses in the character expansion cache tables 21A and 21B are cached. It is stored as management information.

Conventionally, cache management tables 22A, 2
2B, for example, as shown in FIG. 24 (c), the cache management information about the character patterns expanded in the character expansion cache tables 21A and 21B is in the pattern address order [character address order; in FIG. 24 (c), ~
Indicates its address). Then, when the host requests the printing of the character pattern in a predetermined character size, the cache management tables 22A and 22B are based on the character address of the character pattern.
Search the cache management information that has the corresponding letter address (hereinafter also referred to as pattern address) from
When the corresponding one is searched and the character sizes match, the expansion destination address of the cache management information is referred to, and the character pattern stored in the expansion destination address of the character expansion cache tables 21A and 21B is read.

When searching the cache management tables 22A and 22B, first, the intermediate position of the data registered in the cache management tables 22A and 22B up to the present is obtained, and the characters searched up to the middle are searched. The registration position of the desired cache management information is obtained by repeating the procedure of checking whether there is any. For example, an example shown in FIG. 24C will be described. When the cache management information of the addresses ~ is stored in the cache management tables 22A and 22B, and if the pattern address of the character pattern to be searched is, first, the intermediate position of the address ~ becomes the address, and the pattern address Is not between the addresses ~, but between the addresses ~, so when the intermediate position is next obtained for the address ~ side, a pattern address is hit.

The cache management table 22 as described above
In A and 22B, the cache management information is arranged in the order of the character addresses, and accordingly, according to the new character pattern being stored in the character expansion cache tables 21A and 21B, the new cache management information about the new character pattern is cache managed. When registering in the tables 22A and 22B, the data in the cache management tables 22A and 22B are shifted, and new cache management information is inserted and registered so that the character addresses are in order.

For example, as shown in FIG. 24A, while the cache management information is registered in the cache management tables 22A and 22B in the order of the pattern address, ..., The cache management information of the pattern address is registered. 24B, first, as shown in FIG. 24B, the cache management information of the pattern address and the pattern address are shifted to make a gap with the pattern address, and then, as shown in FIG.
As shown in 4 (c), new cache management information of the pattern address is set between the pattern address and the pattern address.

On the other hand, by providing two sets of the character expansion cache tables 21A and 21B and the cache management tables 22A and 22B as described above, when both tables become full, only one table is reset. And switch the registration destination table,
Data about a new character pattern is registered in the table. As a result, the character pattern is always stored in either one of the tables, and a certain level of cache hit rate is ensured.

Such switching operation will be described with reference to FIGS. First, after registering a character pattern and cache management information in the character expansion cache table 21A and the cache management table 22A of one group, when the cache management table 22A of this group becomes full, the character expansion cache of the other group is registered. Table 21A
The character pattern and the cache management information are also registered in the cache management table 22A.

Then, as shown in FIG. 25, when the cache management table 22B also becomes full, both the cache management tables 22A and 22B are full, and new character patterns and cache management information cannot be registered anymore. Therefore, at that point, as shown in FIG. 26, one set of the character expansion cache table 21A and the cache management table 22A is reset (actually, by resetting only the cache management table 22A, the character expansion cache table 21A is reset). Data cannot be read, and the character expansion cache table 21A is also reset), and the character expansion cache table 21A and the cache management table 22A are
New character patterns and cache management information will be registered again.

At this time, since the pipeline processing is performed, the character expansion cache tables 21A, 21B and the cache management table 22A,
22B, each of which is controlled by a CPU different from that of the cache management table 22B.
(Registration control of the character pattern for) is proceeding ahead of.

[0016]

However, in the above-described conventional cache control method for outline character expansion, as described with reference to FIGS. 24A to 24C, in the cache management tables 22A and 22B, each character Since the cache management information about patterns is arranged in the order of pattern addresses, cache control can be performed in a small storage area and the search speed is fast, but it takes a considerable amount of time to register characters due to the processing in order of the pattern addresses, which is inefficient. There is a problem.

Further, as described above, since the pipeline system is adopted, the cache management information registration control and the outline character expansion control are controlled by different CPUs,
In the cache management information registration control, the cache management tables 22A and 22B are immediately restored and registered in the cache management tables 22A and 22B, and there is no need to wait for the completion of the expansion in the outline character expansion control (the completion of registration of the character pattern in the character expansion cache tables 21A and 21B). Therefore, there is a gap between the cache management information registration and the end of outline character expansion control.

When an initialized command is received from the upper CPU to correct such a gap, the cache management table 22A,
The final position of the cache management information registration in 22B is
The character expansion cache tables 21A and 21B must be returned to the same position as the final character expansion position.
At this time, during the table switching operation as described in FIGS. 25 and 26, for example, the data storage state as shown in FIG. 27 immediately after switching (that is, the final character pattern is still registered in the character expansion cache table 21B). With no
When new cache management information is registered in advance in the cache management table 22A) and the initialization command is received, as shown in FIG. 28, the new cache management table 22A that has been registered in the switching destination cache management table 22A is newly registered. All cache management information will be reset.

For this reason, after initialization by the initialized command, the cache hit ratio is temporarily about 1 /.
This is a factor of 2 or less, which is a factor that hinders the speeding up of character processing. Further, in the conventional outline character expansion cache control method, as shown in FIGS. 29 and 30, a character mother cache control unit 23 having a character character cache 23a for storing character character data logged from the host.
When the outline character is expanded based on the character data stored in the character cache 23a in combination with the character data and the character data in the character cache 23a is exchanged from the host side, the following problems occur.

For example, when the character mother data is replaced with the character mother data in the character mother cache 23a, as shown in FIG.
As shown in FIG. 9, the character expansion cache table 21A,
When a character pattern for character data is registered in 21B, cache management information for that character pattern is also registered in the cache management tables 22A and 22B.

Therefore, when there is a print request for character data after data replacement, the cache management table 22
A, 22B is a character output error that the character pattern for the character data is printed in the portion where the character pattern for the character data is printed because the character pattern is searched using the character address and character size of the cache management information as keys. Will occur.

[0022] The present invention has been made in view of such problems, purpose of the present invention, character registration faster than the cache control to manage side by conventional pattern address order (the cache registration) to so enable Te, Ru der to provide an outline character development apparatus that realizes efficient process.

[0023]

[0024]

FIG. 1 is a block diagram of the principle of the present invention. In FIG. 1, 1 is a character expansion cache table, 2 is a cache management table, and 3 is a search code table. Here, the character expansion cache table 1 stores character patterns obtained by expanding outline characters from character data according to a predetermined character size.

Further, the cache management table 2 sequentially stores, as cache management information, character addresses, character sizes, and expansion destination addresses in the character expansion cache table 1 for each character pattern stored in the character expansion cache table 1. To do. further,
The search code table 3 extracts the storage address of each cache management information in the cache management table 2 by extracting a part of the letter address of each cache management information.
It is stored in the obtained search address.

When the host requests the output of a character pattern in a predetermined character size, first, the search code table 3
, A part of the character address of the character pattern is taken.
The storage address stored in the search address obtained by the retrieval is referred to. Then, in the cache management table 2, the character address and the character size of the cache management information stored in the storage address obtained from the search code table 3 are referenced.

After that, if the character address and character size referred to in the cache management table 2 match the character address of the character pattern and the predetermined character size, respectively, refer to the expansion destination address of the cache management information. Then, the character pattern stored in the expansion destination address of the character expansion cache table 1 is read out and controlled (the above is the first claim). At this time, when character patterns having the same search address but different character addresses are registered in the cache management table 2, a next pointer designating the address of the character pattern having different character addresses is used as the cache management information. , Cache management table 2 and cache management table 2
And the character size and character size
The letter address of the turn and the specified character size match.
If not, the next pointer of the cache management information
While referring to it, it is necessary to wait until the next pointer reaches 0.
Character address and sentence referred to in the management table 2
Character size, character address of the character pattern and predetermined
The cache management information that matches the character size
Cache management information that matches whether it exists or not
If exists, the expansion destination ad of the relevant cache management information
Responsive, and the exhibition of the character expansion cache table 1
Read control of the character pattern stored in the groove address
It may be configured as described above (Claim 2).

FIG. 2 is a principle block diagram of another embodiment of the present invention. As shown in FIG. 2, in the second invention, two sets of character expansion cache tables 1A and 1B and cache management tables 2A and 2B are provided. It is provided. Each of the character expansion cache tables 1A and 1B stores a character pattern obtained by expanding outline characters from character data according to a predetermined character size. The cache management tables 2A and 2B respectively store the character patterns. The character expansion addresses and character sizes for the respective character patterns stored in the character expansion cache tables 1A and 1B and the expansion destination addresses in the character expansion cache tables 1A and 1B are stored as cache management information.

Further, 4 is a switching dedicated buffer for temporarily storing new cache management information when the stored data in the two sets of cache management tables 2A and 2B is full, and 5 is two sets of characters. Two sets of cache management tables 2 in the expanded cache tables 1A and 1B
This is an outline character expansion final flag set when all the character patterns corresponding to the cache management information stored in A and 2B are stored.

When the host requests the output of a character pattern of a predetermined character size, first, the cache management table 2 is generated based on the character address of the character pattern.
A and 2B are searched for the cache management information corresponding to the character address of the character pattern, and when the character sizes match, the expansion destination address of the cache management information is referred to, and the character expansion cache table 1
The character pattern stored in the expanded address of A and 1B is read and controlled.

The above two sets of cache management tables 2
When the stored data in A and 2B becomes full, the stored data in one cache management table 2A is reset, and new character patterns and cache management information are stored in one character expansion cache table 1A and cache management table 2A. It is like this.

At this time, in the mode shown in FIG. 2, when the outline character expansion final flag 5 is set, the stored data in one cache management table 2A is reset and temporarily stored in the switching dedicated buffer 4. and it has to be copied to the cache management information in one of the cache management table 2A. 3 is a principle block diagram of another aspect of the present invention. In FIG. 3, 6 is a character cache for storing character data, 1 is an outline character from the character data stored in the character cache 6 according to a predetermined character size. It is a character expansion cache table which stores the character pattern obtained by expanding the character.

2 is a character expansion cache table 1
3 is a cache management table for storing character address, character size, and expansion destination address in the character expansion cache table 1 for each character pattern stored as the cache management information. Further, reference numeral 7 denotes an interface unit for notifying the cache management table 2 and the character expansion cache table 1 of the replacement of character data in the character buffer 6 from the host side.

The even mode smell shown in FIG. 3, when there is an output request of a character pattern at a predetermined character size from the host searches the cache management table 2 based on the alphabet address of the character pattern, of the character pattern When the cache management information corresponding to the character mother address is searched and the character sizes match, the expansion destination address of the cache management information is referred to, and the character pattern stored in the expansion destination address of the character expansion cache table 1. Read control.

At this time, in the mode shown in FIG. 3, when the character data in the character buffer 6 is exchanged from the host side, the cache management table 2 and the character expansion cache table 1 notified from the interface unit 7 are received.
If the cache management information and the character pattern of the character data before replacement is stored in the replaced character data, the cache management information and the character pattern are deleted, and the character data after replacement is deleted from the replaced character data. stores the character pattern obtained by the outline characters developed into character development cache table 1, that stores the cache management information for the character pattern cache management table 2.

[0036]

[Operation]BookInvention outline character expansionapparatus
(Claim 1, 2), The cache management table 2
Cache management information is registered and stored in order, and
The storage address of the cache management information of the
Character address in the cache management information in table 3
SuGet a part of the gillSearch address (letter
Stored in the address obtained by processing the mother address)
There is.

When a character pattern output request is made by the host, a part of the character address of the character pattern is extracted.
The search code table 3 is searched using the search address obtained as a search key, and the storage address stored in the search address is referred to. Then, after that, in the cache management table 2, like the normal processing,
The character address and character size of the cache management information stored in the storage address from the search code table 3 are referenced, and these character address and character size match the character address and character size of the character pattern for which the output request is made. If so, the character pattern stored in the expansion destination address of the cache management information is read out from the character expansion cache table 1. As a result, the character patterns can be searched while sequentially registering and storing the cache management information in the cache management table 2, and there is no need to arrange and manage the character patterns in the conventional pattern address order.

Further, in the outline character expanding device of the aspect shown in FIG. 2 described above, when the stored data in the two sets of cache management tables 2A and 2B becomes full,
Until the final character expansion flag 5 is set,
The new cache management information is temporarily stored in the dedicated buffer 4 at the time of switching. When all the character patterns corresponding to the cache management information stored in the two sets of cache management tables 2A and 2B are stored in the two sets of character expansion cache tables 1A and 1B, the outline character expansion final flag 5 is set. After being set, the stored data in one cache management table 2A is reset, and the cache management information temporarily stored in the switching dedicated buffer 4 is copied to one cache management table 2A, The registration process of is continued. As a result, when the cache management table 2A does not need to be reset even when the initialized command is received, the cache management table 2A can be returned to the original state after the normal reset.

Further, in the outline character expanding device of the aspect shown in FIG. 3, the character buffer 6 is input from the host side.
When the character data in is replaced,
The interface unit 7 notifies the cache management table 2 and the character expansion cache table 1 to that effect. If the cache management table 2 and the character expansion cache table 1 store the cache management information and the character pattern of the character data before the replacement for the replaced character data, the cache management is performed. The information and the character pattern are deleted, the character pattern obtained by expanding the outline character from the replaced character data is stored in the character expansion cache table 1, and the cache management information about the character pattern is stored in the cache management table 2. Is stored.

[0040]

Embodiments of the present invention will be described below with reference to the drawings. (A) Description of a system to which the present invention is applied FIG. 4 is a block diagram showing a configuration example of a system to which the present invention is applied. In FIG. 4, 100 is a character pattern output request in a predetermined character size or a character to be described later. A host for exchanging character data in the cache, 101 is a channel unit for receiving data from the host 100, 10
Reference numeral 2 is a data analysis unit (DTP) that analyzes and outputs data from the host 100, and 103 is a device control unit that receives the analysis result from the data analysis unit 102 and controls the entire device (system) according to a command from the host 100 ( CTP).

An order issuing unit (I) 104 further analyzes the data received from the device control unit 103 and creates and sends an order for causing an IHP, which will be described later, to perform drawing.
CP), 105 is a drawing processing unit (IH) for drawing the character pattern requested by the host 100 for output (printing) in the requested character size in response to the order from the ICP 104.
In P), this IHP 105 is composed of an outline character expansion unit 106 and a drawing unit 107.

Here, the outline character expansion unit 106
Of the character pattern requested by the host 100 for output this time, if the requested character size is not registered in the character expansion cache table described later, the outline character is expanded from the character data according to the predetermined character size. The character pattern is created. The drawing unit 107 draws the character pattern expanded by the outline character expansion unit 106 or the character pattern read from the character expansion cache table on a bitmap memory (not shown).

Further, 108 is an output unit for printing out a character pattern from the IHP 105, and 109 is an ICP.
A command analysis unit for analyzing an order (command) from 104, a character processing unit 110 for processing a character, and a reference numeral 111 for registering a character pattern expanded by the IHP 105 in a character expansion cache table described later while expanding the character. A character pattern cache control unit 112 that searches for and controls reading of a character pattern registered in the cache table returns a cache registration pointer to a position where the IHP 105 has actually completed the expansion when receiving an initialization command (initialized command). A cache reset unit that performs cache reset processing.

In the system configured as described above, a pipeline system for performing various kinds of processing in parallel is adopted, and the character pattern cache control unit 111 causes the IH
The character pattern raster-expanded from the outline character in P105 is registered in a character expansion cache table described later, and the character pattern already raster-expanded once can be reused to speed up character processing.

[0045] The onset Ming, relates to a cache control according to the character pattern cache control unit 111, in the following embodiments, a character pattern cache control unit 111
The cache control such as character pattern search / registration in the character expansion cache table and the cache management table will be described in detail. (B) Description of the First Embodiment FIG. 5 is a block diagram showing the main structure of an outline character expansion device as the first embodiment of the present invention. As shown in FIG. Character expansion cache tables 1A and 1B and two sets of cache management tables 2
A, 2B, two sets of search code tables 3A, 3B,
Two sets of registration number pointers 8A and 8B are provided. In this embodiment, those marked with "A" are primary
Called cache, with a "B" attached Secondary
Call it cache.

Here, the character expansion cache tables (outline character expansion areas) 1A and 1B each store a character pattern obtained by expanding outline characters from character data in accordance with a predetermined character size. The cache management tables 2A and 2B are sequentially stored as cache management information for each character pattern stored in the character expansion cache tables 1A and 1B, respectively.

As the cache management information, as shown in FIG. 5, a character address, a character size, and an expansion destination address in the character expansion cache tables 1A, 1B for each character pattern stored in each character expansion cache table 1A, 1B. In addition to the above, this embodiment has a next pointer. This next pointer is cache management information about a character pattern when another character pattern having the same search code (search key) described later but different information (character address, character address) is registered. Specifies the next address of the cache management tables 2A and 2B registered / stored therein, and its specific function will be described later in detail with reference to FIG.

Further, the search code tables 3A and 3B are
The storage address of each cache management information in the cache management tables 2A and 2B is registered and stored in the search address set according to the letter address of each cache management information. The search address (search code, also referred to as a search key) is, in this embodiment, is processed by the outline character pattern address (Jamo address), and scores eject the part.
For example, as shown in FIG. 6, in the case of an 8-byte pattern address, the middle 4 bytes are taken out as a search key. In this embodiment, the search key thus obtained is used as the offset position (search address) of the search code tables 3A and 3B.

Further, the registration number pointers 8A and 8B indicate the location (storage address, offset position) of the cache management information to be registered next in the cache management tables 2A and 2B, respectively. Since the cache management information is sequentially registered in 2B, the range in use of each cache management table 2A, 2B can be determined by referring to the registration number of the registration number pointers 8A, 8B.

By setting the registered number pointers 8A and 8B to 0, the character expansion cache table 1
A, 1B, cache management tables 2A, 2B, search code tables 3A, 3B can be set without changing the contents of the cache management tables 2A, 2B, that is, in an unused area without changing the contents. Yes, this allows a cache reset.

As shown in FIG. 5, the overall cache configuration of this embodiment includes a primary cache and a secondary cache, and when both caches become full, one cache is reset. Therefore, the character pattern is always registered / existed in one of the caches. Specifically, it is first registered in the primary cache, and when the primary cache becomes full, it is registered in the secondary cache. When the secondary cache becomes full, both caches become full and cannot be registered. Therefore, when registering the next character, reset the primary cache and register again. After that, when the primary cache is full, the secondary cache is reset. When both caches are full, the cache to be used next is reset and re-registered. The cache reset at this time is performed by setting the registration number pointer 8A or 8B to 0 as described above.

There are the following two conditions for judging that each cache is full. (1) When the cache management tables 2A and 2B are full (2) When there is no area that can be expanded / registered / stored in the character expansion cache tables (outline character expansion areas) 1A and 1B, as shown in FIG. As described above, the search key (search code, search address) of the present embodiment is obtained by extracting a part of the outline character pattern address (character address). In the present embodiment, the search code tables 3A and 3B are created as a factor for referring to the cache management tables 2A and 2B with this search key.

Since the search code tables 3A and 3B must support all the search keys, they support 0000 to FFFF (hexadecimal number), for example, as shown in FIG. Further, the search code tables 3A and 3B store the offset position of the cache management table corresponding to the search key.

Then, the cache management tables 2A and 2B at the designated offset positions are referred to, and it is determined whether the cache management information in the cache management tables 2A and 2B is of the character pattern being searched. At this time, the processing is performed under the following conditions. (1) It is determined whether the designated cache management table 2A or 2B is within the range in use.

The judgment within the range in use is made by the registration number pointer 8A or 8B (each registration number pointer 8
A and 8B are cache management tables 2 to be registered next.
Offset positions of A and 2B are set). That is,
If it is in use, the cache management information in the cache management table 2A or 2B is referred to. If it is not used, it is determined that the cache management information is not registered in the cache management table 2A or 2B, and the cache is reset. Since it is the information before it is registered, the searched character pattern is registered in the cache.

(2) A search key is extracted from the pattern address on the cache management information in the specified cache management table 2A or 2B in the same manner as above, and the search key extracted from the pattern address currently searched for. Compare with. If the search keys match, the cache management information in the cache management table 2A or 2B is referenced, while if the search keys do not match,
Since it is the information before the cache reset, the character pattern being searched for is registered in the cache.

As described above, the cache management table 2A
Alternatively, by performing a reverse search from 2B, the search code table 3A or 3B can be used without reset processing even if the cache is reset. In the above (1) and (2), if it is found that the cache management information is registered in the cache management table 2A or 2B, the pattern on the cache management information in the designated cache management table 2A or 2B. Compare the address and pattern size (character size) with the pattern address and character size of the character pattern being searched.

At this time, when the search keys are the same but the pattern address and the character size are different, the next pointer on the cache management information in the cache management table 2A or 2B is referred to until the next pointer becomes 0.
It is determined whether or not there are patterns with the same pattern address and character size. If there is the same pattern address and character size before the next pointer becomes 0, the rasterized character pattern of the character pattern being searched for is registered in the character expansion cache table 1A or 1B. Therefore, the character pattern is read based on the expansion destination address.

On the other hand, when it is not possible to retrieve the pattern address and the character size which are equal by the time the next pointer becomes 0, the character expansion cache table 1A or 1B has not yet retrieved the character pattern being retrieved. Since the character pattern is not registered, IHP1
At 05, the outline character is raster-expanded, and the result is newly expanded / registered / stored.

Here, the above-mentioned next pointer will be specifically described with reference to FIG. In FIG. 7, in the cache management table 2A (2B), for example, the cache management information about the character pattern of the pattern address “FF000100” is stored in the address, and the pattern address “FF00” is stored in the address.
It is assumed that the cache management information about the character pattern of "0111" is stored.

Since the search keys for these character patterns are all "0001", the next pointer of the cache management information at the previously stored address stores the next cache management information having the same search key. The address is registered and set. Further, if cache management information having the same search key is not stored after the cache management information at the address, 0 is registered and set in the next pointer of the cache management information at the address.

Then, the search code table 3A (3B)
The storage address (offset position) of the cache management information previously registered in the cache management table 2A (2B) is registered and stored in the address corresponding to the search key "0001". Although the case where the storage keys are the same and the pattern addresses are different has been described here, the next pointer is similarly set when the storage keys are the same and the character size (pattern size) is different.

Next, the cache control method according to the present embodiment described above will be described in detail with reference to the flowcharts shown in FIGS. 8 to 11. First, as shown in FIG. 8, when a print request is received from the host 100 ( In step A1), the request from the host 100 is analyzed, and a development request and a print request for each character are made (step A2). An arbitrary part (in the case of the present embodiment, the middle 4 bytes shown in FIG. 6) of the character address of the character pattern for which printing is requested is extracted and used as the search code (search key) (step A).
3).

Then, it is judged whether or not the cache management information corresponding to the requested retrieval code of the character pattern is registered in the cache management tables 2A and 2B (step A4). The character pattern is expanded into outline characters and the cache management information is registered at the offset position indicated by the registration number pointer 8A (8B) according to the procedure described later with reference to 10, and corresponds to the search code in the search code table 3A (3B). The offset position (stored address) is registered in the address (step A5).

[0065] The step A4, if it is determined to be registered refers to the cache management table 2A (2B) cache management information stored in the address corresponding to the retrieval code (Step A 6), It is determined whether or not the search code of the requested character pattern and the search code generated from the letter address in the cache management information referred to in the cache management table 2A (2B) are the same (step A7). If they are not equal to each other, the character pattern is expanded into outline characters, the cache management information is registered at the offset position indicated by the registration number pointer 8A (8B), and the search code table 3A (3B), just as in step A5 described above. The offset position (stored address) is registered in the address corresponding to the search code of () (step A8).

If it is determined in step A7 that they are equal, the process proceeds to the flowchart of FIG. 9 and the character address of the requested character pattern and the character in the cache management information referenced in the cache management table 2A (2B). It is determined whether the address and the address are the same (step A
9) If they are equal, it is determined whether the character size of the requested character pattern is equal to the character size in the cache management information referenced in the cache management table 2A (2B) (step A10).

If it is determined in step A10 that they are equal, the requested character pattern is the same as the character pattern corresponding to the cache management information referenced in the cache management table 2A (2B). Cache management table 2A (2
The output destination (printing unit) 108 is requested to print by designating the expansion destination address set in B) (step A1).
4) The character pattern (raster pattern) expanded and stored at the specified expansion destination address is read from the character expansion cache table 1A (1B) and printing is executed (step A15).

On the other hand, step A9 or step A1
When it is determined that they are not equal in 0, it is determined whether or not the next pointer set in the referred cache management information is other than 0 (step A11). If it is determined to be other than 0, the cache management information pointed to by the next pointer is referred to (step A12), and the determination in steps A9 and A10 described above is executed.

If it is determined to be 0 in step A11, the character pattern is expanded into outline characters and the cache management information is set to the offset position indicated by the registration number pointer 8A (8B) by the procedure described later with reference to FIG. , But in this case, search code table 3
A (3B) is not registered, and the next point in the last-referenced cache management information is set from 0 to the storage address at which the newly registered cache management information is registered, that is, the offset position indicated by the registration number pointer 8A (8B). (Step A13).

If NO is determined in step A4 or A7 in FIG. 8, the process shown in FIG. 10 is executed as described above. That is, the printing process is performed after the information about the character pattern being searched is registered in the search code table 3A (3B), the cache management table 2A (2B), and the character expansion cache table 1A (2A).

First, the storage address of the cache management table 2A (2B) for newly registering cache management information (the next pointer is 0) at the address corresponding to the search code in the search code table 3A (3B) [registration number pointer 8
Offset position indicated by A (8B)] is set (step A16). Then, the cache management information about the new character pattern is stored in the cache management table 2A (2
B) is registered and set (step A17), and the character expansion cache memory 1A (1
A character is expanded at the expansion destination address of B) to generate a character pattern (raster pattern) (step A18).

Thereafter, as in steps A14 and A15, the expansion destination address is designated and the output section (print section) 1
08 (step A14), the character pattern (raster pattern) expanded and stored at the specified expansion destination address is transferred to the character expansion cache table 1
Read from A (1B) and execute printing (step A
15).

In step A11 of FIG. 9, NO
When the determination is made, the processing shown in FIG. 11 is executed as described above. That is, the printing process is performed after the information about the character pattern being searched for is registered in the cache management table 2A (2B) and the character expansion cache table 1A (2A). First of all, the next pointer in the cache management information referred Finally, from 0,
After changing the storage address where the newly registered cache management information is registered, that is, the offset position indicated by the registration number pointer 8A (8B) (step A21), the processes of steps A22 to A25 are executed. Here, the processing in steps A22 to A25 is exactly the same as the processing in steps 17 to A20 in FIG. 10, and thus the description thereof will be omitted.

As described above, according to the first embodiment of the present invention, by extracting a part from the pattern address of the outline character and using it as a search key, cache control is performed, and the cache is arranged and managed in the conventional pattern address order. Character registration (cache registration) can be performed faster than control, and efficient processing can be realized.

Generally, since outline character expansion takes time, speeding up the cache registration performed during outline character expansion has a great effect on character output and improves the performance of printing devices such as printers and other display devices. It will make a big contribution. (C) Description of Second Embodiment FIG. 12 is a block diagram showing the main configuration of an outline character expanding device as a second embodiment of the present invention. As shown in FIG. Similar to the embodiment, two sets of character expansion cache tables 1A and 1B and two sets of cache management tables 2A and 2B are provided. Although not shown in the figure, two sets of search code tables 3A and 3B similar to those of the first embodiment are provided.

In the second embodiment, as shown in FIG. 12, the character pattern cache control unit 111 is provided with the switching dedicated buffer 4 and the outline character expansion final flag 5. Here, the switching dedicated buffer 4 temporarily stores new cache management information when the stored data in the two sets of cache management tables 2A and 2B becomes full, and outline character expansion is performed. The final flag 5 is set to "1" when all the character patterns corresponding to the cache management information stored in the two sets of cache management tables 2A and 2B are stored in the two sets of character expansion cache tables 1A and 1B. It is what is done.

Now, as shown in FIG. 12, the overall cache structure of this embodiment has a primary cache and a secondary cache as in the first embodiment.
When both caches are full, reset one cache. Specifically, when the primary cache is registered first, and then the primary cache becomes full, it is registered in the secondary cache, and when the secondary cache becomes full, both caches become full and cannot be registered. , When registering the next character, reset the primary cache and register again. After that, when the primary cache is full, the secondary cache is reset. When both caches are full, the cache to be used next is reset and re-registered. It is assumed that the cache management tables 2A and 2B are full.

Next, the operation of this embodiment will be described with reference to FIGS. First, the procedure for switching from the cache management table 2B full to the cache management table 2A will be described with reference to FIGS. As shown in FIG. 12, the cache management table 2A
Suppose that the cache management information is registered in the cache management table 2B in a full state, the cache management table 2B also becomes full. At this time, the character pattern expanded / registered / stored in the character expansion cache table 1B is performed later than the cache management information registered in the cache management table 2B.

When the cache management table 2B becomes full, as shown in FIG. 13, the cache management table 2A is not immediately reset, but a new cache to be registered in the cache management table 2A is actually used. The management information is registered in the dedicated buffer 4 at the time of switching. Then, until the outline character expansion final flag 5 is set, the new cache management information is stored in the switching dedicated buffer 4
Will be temporarily registered and stored in. As shown in FIG. 14, when all the character patterns corresponding to the cache management information stored in the cache management table 2B are stored in the character expansion cache table 1B, the outline character expansion final flag 5 is set to "1". To be done.

When "1" is set in the outline character expansion final flag 5, the stored data in the cache management table 2A is reset, as shown in FIG. By resetting the cache management table 2A, the data of the character expansion cache table 1A cannot be read, and the character expansion cache table 1A is also reset. Then, after copying the cache management information temporarily stored in the switching dedicated buffer 4 to the cache management table 2A,
Normal registration processing continues.

For example, in the state shown in FIG. 13, that is, in the table switching operation, the final character pattern is not yet registered in the character expansion cache table 1B, and new cache management information is registered in the switching dedicated buffer 4. When the initialization command is received while the cache management table 2B is switched to another in the cache management table 2B in this embodiment, the cache management table 1A of the switching destination is not reset as shown in FIG. Only the cache management information that precedes the outline character expansion in the hour dedicated buffer 4 is reset.

Therefore, if the cache management table 2A does not need to be reset even when the initialized command is received, the cache management table 2A can be held in the original state after the normal reset, and the cache management table 2A can be temporarily stored as in the conventional case. Moreover, the cache hit rate will not be reduced to less than about 1/2. Next, the cache registration control and the initialization control upon receiving the initialized command according to the present embodiment will be described with reference to FIGS.

FIG. 17 shows a case where the area where the IHP has actually completed the expansion has not moved to the next cache (secondary cache). In the ICP, the registered cache becomes the secondary cache. After moving, IH
The cache management information is registered in the dedicated buffer 4 at the time of switching until the expansion of P is caught up. In addition, FIG.
This shows a case where the area actually expanded by P is moved to the next cache (secondary cache). Note that the contents of the switching dedicated buffer 4 are copied to the cache management table 2B at the time when the area where the IHP has actually been expanded is moved to the next cache (secondary cache).

FIG. 19 is a flow chart for explaining the cache registration control procedure. As shown in FIG. 19, the expanded outline characters are sequentially registered in the cache, and the cache in which the cache management information should be registered is the next cache. When moving to (step B1), IHP
It is determined whether or not the end area of the outline character expansion in (4) is moved to the next cache (that is, whether the outline character expansion final flag 5 is set) (step B).
2).

At this time, if it is determined that the end area of the outline character expansion in IHP has not moved to the next cache (the outline character expansion final flag 5 is not set), then FIG. 13 or FIG. As shown in, the cache management information is temporarily registered in the switching dedicated buffer 4 (step B3). Then, when it is determined that the end area of the outline character expansion in the IHP has moved to the next cache (the outline character expansion final flag 5 is set), it is temporarily registered in the switching dedicated buffer 4. The cache management information is copied to the formal cache (the destination cache management table 2A or 2B) (step B4), and subsequently expanded outline characters are registered in the cache as shown in FIGS. 15 and 18. B5).

When the initialization command (initialized command) is received, the cache state is returned to the position where the actual outline character expansion is performed. At this time, in this embodiment, as described above, by registering the cache management information in the switching dedicated buffer 4 at the time of switching, it is not necessary to invalidate the cache of the moving destination when the initialization command is received, The cache hit rate (cache efficiency) can be improved.

FIG. 20 is a flow chart for explaining the initialization control procedure at the time of receiving the initialized command. As shown in FIG. 20, when the initialization command is received, the IHP actually finishes the expansion. The registration pointer of the ICP is returned to the existing pointer (step C1). Then, a pointer whose next pointer registered in the cache management table 2A or 2B is larger than the pointer whose expansion is actually completed by IHP is deleted (step C2).

Then, an offset of the expansion destination address registered in the cache management table 2A or 2B is larger than the pointer that has actually completed the expansion by the IHP is calculated (step C3), and the cache search table (FIG. 5). The offset of the cache management table 2A or 2B registered in the reference numerals 3A and 3B) is
The initialization processing is completed by deleting the offset larger than the offset obtained in step C3 (step C).
4).

As described above, according to the second embodiment of the present invention, when the stored data in the two sets of cache management tables 2A and 2B becomes full, new data is added until the outline character expansion final flag 5 is set. Cache management information is temporarily stored in the dedicated buffer 4 at the time of switching, and the stored data in one of the cache management tables 2A or 2B is reset after setting the outline character expansion final flag 5, and the cache management in the dedicated buffer 4 at the time of switching is performed. One of the cache management tables 2A or 2
Since the normal registration processing is continued after copying to B, the cache management table 2A or 2B can be strictly maintained if the cache management table 2A or 2B does not need to be reset even when the initialized command is received. It is possible to prevent the cache hit rate from decreasing.

(D) Description of Third Embodiment FIG. 21 is a block diagram showing the essential structure of an outline character expanding device as a third embodiment of the present invention. As shown in FIG. Similarly to the first embodiment, two sets of character expansion cache tables 1A and 1B and two sets of cache management tables 2A and 2B are provided, and although not shown in the drawing, similar to the first embodiment. Two sets of search code tables 3A and 3B are provided.

In the third embodiment, as shown in FIG. 21, a character mother cache control unit 9 having a character character cache 6 for storing character character data logged from the host 100 is combined and stored in the character character cache 6. Outline characters are expanded based on the generated character data. Here, in the present embodiment, as shown in FIG. 21, the interface unit 7 is provided between the character mother cache control unit 9 and the character pattern cache control unit 111.

The interface unit 7 is provided by the host 10
When the character data in the character buffer 6 is exchanged from the 0 side, the fact is notified to the cache management tables 2A and 2B and the character expansion cache tables 1A and 1B of the character pattern cache control unit 111. Therefore, in this third embodiment, the host 1
When character data is replaced in the character buffer 6 from the 00 side, for example, when character data is replaced in the character cache 6 in the character cache 6,
First, the interface unit 7 notifies the replacement information to the cache management tables 2A and 2B and the character expansion cache tables 1A and 1B of the character pattern cache control unit 111.

These cache management tables 2A, 2
B and the character expansion cache tables 1A and 1B store cache management information and character patterns of the character data before replacement for the character data that has been replaced, as shown in FIG. The cache management information and the character pattern in the tables 2A and 2B and the character expansion cache tables 1A and 1B are deleted, and the newly expanded character is prepared to be registered in the cache.

Then, as shown in FIG. 23, the character pattern obtained by expanding the outline character from the replaced character data is stored in the character expansion cache tables 1A and 1B, and the cache management information about the character pattern is stored. Are stored in the cache management tables 2A and 2B. As described above, according to the third embodiment of the present invention, when the character data in the character buffer 6 is exchanged from the host 100 side, the interface unit 7 notifies the cache management tables 2A, 2 to that effect.
B and the character expansion cache tables 1A and 1B are notified, and even if the character replacement in the character cache 6 occurs, the corresponding processing can be performed and the character output error of the character pattern can be surely performed. It can be prevented.

[0095]

As described above in detail, according to the outline character expanding device of the present invention (claims 1 and 2) , the search code created by processing the letter address of the outline character is used in the search code table. By managing the storage address of the cache management information in the cache management table, the cache management information can be registered in order in the cache management table, which is much faster than the conventional cache control in which the cache character management addresses are arranged in order. Character registration can be performed, efficient processing can be realized, and there is an effect that it greatly contributes to the performance improvement of the printing device and the display device.

Further, according to the outline character expansion device of another aspect of the present invention, when the stored data in the two sets of cache management tables becomes full, a new cache character is added until the outline character expansion final flag is set. Temporarily stores the management information in the special buffer for switching, resets the data stored in one cache management table after setting the outline character expansion final flag, and copies the cache management information in the special buffer for switching to one cache management table As a result, if the cache management table 2A or 2B does not need to be reset even when the initialized command is received, the cache management table can be strictly maintained, and the cache hit rate can be prevented from lowering.

Further, according to the outline character expansion device of another aspect of the present invention, when the character data in the character buffer is exchanged from the host side, the interface unit informs the effect to that effect. Since the notification is made to the cache table, even if the character replacement in the character cache occurs, it is possible to perform a process corresponding to this, and it is possible to reliably prevent a character output error in the character pattern.

[Brief description of drawings]

FIG. 1 is a principle block diagram of the present invention.

FIG. 2 is a principle block diagram of another aspect of the present invention.

FIG. 3 is a principle block diagram of another embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration example of a system to which the present invention is applied.

FIG. 5 is a block diagram showing a main configuration of an outline character expanding device as a first embodiment of the present invention.

FIG. 6 is a diagram for explaining a search code according to the first embodiment.

FIG. 7 is a diagram for explaining a function of a next pointer according to the first embodiment.

FIG. 8 is a flowchart for explaining the operation of the first embodiment.

FIG. 9 is a flowchart for explaining the operation of the first embodiment.

FIG. 10 is a flowchart for explaining the operation of the first embodiment.

FIG. 11 is a flowchart for explaining the operation of the first embodiment.

FIG. 12 is a block diagram showing a main configuration of an outline character expanding device as a second embodiment of the present invention.

FIG. 13 is a block diagram for explaining the operation of the second embodiment.

FIG. 14 is a block diagram for explaining the operation of the second embodiment.

FIG. 15 is a block diagram for explaining the operation of the second embodiment.

FIG. 16 is a block diagram for explaining the operation of the second embodiment.

FIG. 17 is a diagram for explaining the operation of the second embodiment.

FIG. 18 is a diagram for explaining the operation of the second embodiment.

FIG. 19 is a flowchart for explaining the operation of the second embodiment.

FIG. 20 is a flow chart for explaining the operation of the second embodiment.

FIG. 21 is a block diagram showing a main configuration of an outline character expanding device as a third embodiment of the present invention.

FIG. 22 is a block diagram for explaining the operation of the third embodiment.

FIG. 23 is a block diagram for explaining the operation of the third embodiment.

FIGS. 24A to 24C are diagrams for explaining a registration state and a registration procedure of cache management information in a conventional cache management table.

FIG. 25 is a diagram for explaining a registration / reset procedure in a conventional cache management table.

FIG. 26 is a diagram illustrating a registration / reset procedure in a conventional cache management table.

FIG. 27 is a diagram for explaining a registration / reset procedure in a conventional cache management table.

FIG. 28 is a diagram for explaining a registration / reset procedure in a conventional cache management table.

[Fig. 29] Fig. 29 is a diagram for describing an operation when replacing a character in a conventional character cache.

[Fig. 30] Fig. 30 is a diagram for explaining an action when replacing a character in a conventional character cache.

[Explanation of symbols]

1,1A, 1B Character expansion cache table 2, 2A, 2B cache management table 3,3A, 3B search code table 4 Dedicated buffer when switching 5 Outline character expansion final flag 6-character mother cash 7 Interface section 8A, 8B registered number pointer 9-character mother cache controller 100 hosts 101 channel 102 Data analysis unit 103 Device control unit 104 Order Placer (ICP) 105 IHP 106 Outline character expansion part 107 Drawing unit 108 Output Section (Printing Section) 109 Command parser 110 Character processing part 111 cache control unit 112 cache reset section

─────────────────────────────────────────────────── ─── Continuation of front page (56) References JP-A-2-238965 (JP, A) JP-A-3-48286 (JP, A) JP-A-6-308932 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) G09G 5/24 B41J 2/485 B41J 5/44 G06F 12/08 G06T 11/20

Claims (2)

(57) [Claims] And 1. A character development cache table storing character pattern obtained by developing depending from alphabet data outline character to a predetermined character size for each character pattern stored in said character development cache tables the alphabet address the character size and expansion destination address definitive in the character development cache tables of the cache management information, and cache management te <br/> over b le for sequentially storing, each cache definitive to the cache management table the storage address of the management information, includes a search code table for storing the search address which is obtained by taking out a part of the alphabet address with each cache management information, there is an output request of the character pattern in a predetermined character size from the host If, Oite to the search code tables, Jamo add of the character pattern Referring to storage address stored in the search for the address obtained by taking out a part of the less, Oite to the cache management table, in the search Kodote <br/> b le or we obtained the storage address Referring to Jamo address and the character size of the cache management information stored, and Jamo address and the character size and hand reference to the cache management table, and the Jamo address and the predetermined character size of the character pattern matches each If
Referring to expand address of the cache management information is configured to read control a character pattern stored in the deployment destination address of the character development cache tables
An outline character expansion device characterized by the above. 2. When character patterns having the same search address but different character addresses are registered in the cache management table, a next pointer designating the address of the character pattern having different character addresses is as cache management information, stored in the cache management table, Jamo address your referred by the cache management table
And character size and character address of the character pattern and
And the specified character size do not match,
While referring to the next pointer of the cache management information,
The cache management table until the counter becomes 0.
Character address and character size that is compared with the character pattern
Character address and the predetermined character size
Determine whether there is matching cache management information
If the matching cache management information exists, the cache
See the expansion destination address of the cache management information and expand the character
Stored at the relevant expansion destination address in the cache table
The outline character expansion device according to claim 1, wherein the outline character expansion device is configured to read and control a character pattern .