## Summary
Fix bug where `AllAndCount(true)` with multiple fields generates invalid
SQL `COUNT(field1, field2, ...)` causing syntax error.
## Root Cause
When `useFieldForCount=true`, the COUNT query inherits the fields
configuration from the model:
```go
// Before (buggy code)
if !useFieldForCount {
countModel.fields = []any{Raw("1")}
}
// When useFieldForCount=true, fields remain as ["id", "nickname"]
// Generates: SELECT COUNT(id, nickname) FROM table ❌
```
## Fix
Always use `COUNT(1)` regardless of `useFieldForCount` parameter since
COUNT() accepts only one argument:
```go
// After (fixed code)
// Always use COUNT(1) for counting, regardless of useFieldForCount.
// COUNT() accepts only one argument, so we can't use multiple fields.
countModel.fields = []any{Raw("1")}
```
Applied to both `AllAndCount()` and `ScanAndCount()` methods.
## Tests
Added `Test_Issue4698` with 5 test cases:
1. AllAndCount(true) with multiple fields
2. AllAndCount(false) with multiple fields (baseline)
3. ScanAndCount with multiple fields
4. AllAndCount with single field
5. AllAndCount with WHERE condition
All tests verify that COUNT generates valid SQL and returns correct
count.
## Related
Fixes#4698
Ref #4703 (discovered during pagination test development)
## Summary
- Fixed a bug where `Raw()` with `Where()` and
`Count()`/`ScanAndCount()` was ignoring the Where conditions in Count
queries
- The issue was in `getFormattedSqlAndArgs()` which returned `nil` for
`conditionArgs` without calling `formatCondition()` for Raw SQL in
`SelectTypeCount` case
## Changes
- Modified `database/gdb/gdb_model_select.go` to call
`formatCondition()` for Raw SQL Count queries
- Added comprehensive test cases for MySQL and PostgreSQL drivers
- Fixed incorrect test expectation in `Test_Model_Raw`
## Test plan
- [x] Added `Test_Issue4500` with 6 edge cases covering:
- Raw SQL with WHERE + external Where condition
- Raw SQL without WHERE + external Where condition
- Raw + Where + ScanAndCount
- Raw + multiple Where conditions
- Raw SQL with no external Where (baseline)
- Verify All() still works correctly
- [x] All tests pass on PostgreSQL
Closes#4500
```golang
func main() {
adapter := gcache.NewAdapterRedis(g.Redis())
g.DB().GetCache().SetAdapter(adapter)
result, count, err := g.Model("TBL_USER").Cache(gdb.CacheOption{
Duration: 100 * time.Minute,
Name: "VIP",
}).AllAndCount(false)
g.DumpJson(result)
fmt.Println(count, err)
}
```
执行这段查询后`g.DumpJson(result)`的结果是`[
{
"COUNT(1)": 5
}
]`,但是正确结果应该是五条用户信息，查看源代码后发现先执行的count查询和后来select查询都是直接使用了`VIP`这个缓存key，在redis中实际缓存key是`SelectCache:VIP`，第二步查询select获得的是count查询的缓存，所以查询结果是错的。
因此为`Model`增加一个`PageCache`方法允许用户分别设置`count query`和`data query`的缓存参数
```golang
// PageCache sets the cache feature for pagination queries. It allows to configure
// separate cache options for count query and data query in pagination.
//
// Note that, the cache feature is disabled if the model is performing select statement
// on a transaction.
func (m *Model) PageCache(countOption CacheOption, dataOption CacheOption) *Model {
model := m.getModel()
model.pageCacheOption = []CacheOption{countOption, dataOption}
model.cacheEnabled = true
return model
}
```
然后`AllAndCount`在查询时分别给两个查询设置对应的缓存参数`ScanAndCount`同理
```golang
// AllAndCount retrieves all records and the total count of records from the model.
// If useFieldForCount is true, it will use the fields specified in the model for counting;
// otherwise, it will use a constant value of 1 for counting.
// It returns the result as a slice of records, the total count of records, and an error if any.
// The where parameter is an optional list of conditions to use when retrieving records.
//
// Example:
//
// var model Model
// var result Result
// var count int
// where := []any{"name = ?", "John"}
// result, count, err := model.AllAndCount(true)
// if err != nil {
// // Handle error.
// }
// fmt.Println(result, count)
func (m *Model) AllAndCount(useFieldForCount bool) (result Result, totalCount int, err error) {
// Clone the model for counting
countModel := m.Clone()
// If useFieldForCount is false, set the fields to a constant value of 1 for counting
if !useFieldForCount {
countModel.fields = []any{Raw("1")}
}
if len(m.pageCacheOption) > 0 {
countModel = countModel.Cache(m.pageCacheOption[0])
}
// Get the total count of records
totalCount, err = countModel.Count()
if err != nil {
return
}
// If the total count is 0, there are no records to retrieve, so return early
if totalCount == 0 {
return
}
resultModel := m.Clone()
if len(m.pageCacheOption) > 1 {
resultModel = resultModel.Cache(m.pageCacheOption[1])
}
// Retrieve all records
result, err = resultModel.doGetAll(m.GetCtx(), SelectTypeDefault, false)
return
}
```
---------
Co-authored-by: houseme <housemecn@gmail.com>
## What this PR does
Revert the auto table prefix behavior in `Order()` and `Group()`
introduced by #4521.
## Why
PR #4521 attempted to resolve column ambiguity in GROUP BY/ORDER BY with
MySQL JOIN by automatically adding table prefixes to unqualified
columns. However,
this approach has issues:
1. When using `.As()` to set table alias, it uses the original table
name instead of the alias, causing errors in PostgreSQL and other
databases
2. The framework cannot reliably determine which table the user intends
when multiple tables have the same column
3. Adds hidden behavior that users may not expect
## Example of the issue (#4554)
```go
db.Model("demo_a").As("a").
LeftJoin("demo_b", "b", "a.id=b.data_id").
Order("sort").All()
Expected (v2.9.5):
ORDER BY "sort"
Actual (v2.9.6):
ORDER BY "demo_a".sort -- Wrong! Should use alias "a" or no prefix
Solution
Revert to v2.9.5 behavior: Order("sort") generates ORDER BY "sort"
without auto-prefixing. Users should explicitly specify table prefix
when needed:
Order("a.sort").
Closes#4554
```
---------
Co-authored-by: hailaz <739476267@qq.com>
本次PR主要针对GoFrame ORM中的软删除`SoftTime`功能进行了优化
- 新增`SoftTimeFieldType`枚举类型，用于区分创建、更新、删除三种不同的软时间字段
- 替代之前使用的魔数方式，提高类型安全性
- 将原有的6个方法精简为4个方法, 合并了三个几乎相同的`GetFieldNameAndTypeFor*`方法为统一的方法
This pull request refactors and simplifies the "soft time"
(created/updated/deleted timestamp) handling logic in the database
layer, making the codebase more maintainable and extensible. The changes
consolidate multiple similar methods into general-purpose ones, improve
cache key generation, and clarify the logic for generating and applying
soft time field values and conditions.
Key changes include:
**Soft Time API Refactoring and Simplification**
- Consolidated multiple methods (`GetFieldNameAndTypeForCreate`,
`GetFieldNameAndTypeForUpdate`, `GetFieldNameAndTypeForDelete`) into a
single, parameterized method `GetFieldInfo`, reducing code duplication
and making it easier to support new soft time field types. The interface
and implementation for soft time maintenance (`iSoftTimeMaintainer`,
`softTimeMaintainer`) have been updated accordingly.
[[1]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L46-R66)
[[2]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L105-R180)
- Combined and renamed methods for generating soft time field values and
delete conditions, such as merging
`GetValueByFieldTypeForCreateOrUpdate` into `GetFieldValue`, and
`GetWhereConditionForDelete` into `GetDeleteCondition`. Related usages
throughout the codebase have been updated to use the new methods.
[[1]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L255-R206)
[[2]](diffhunk://#diff-97beb485550e4381182a04bbb857a25b7f4ecd4a594dff8ac884cfaae38f3046L34-R35)
[[3]](diffhunk://#diff-97beb485550e4381182a04bbb857a25b7f4ecd4a594dff8ac884cfaae38f3046L55-R55)
[[4]](diffhunk://#diff-88304ddb7791aedbd83dafb68374aecab286d1356a7f2f149a8e57ac1a7f40b4L265-R267)
[[5]](diffhunk://#diff-88304ddb7791aedbd83dafb68374aecab286d1356a7f2f149a8e57ac1a7f40b4L298-R311)
[[6]](diffhunk://#diff-d4f6e0370e049dea52f3db9a13c64e2cfb2f7ef012433186e21179149b626d0fL944-R944)
**Soft Delete Logic Improvements**
- Refactored the logic for building soft delete WHERE conditions and
generating update data, with clearer and more robust handling of field
types and prefixes. Introduced helper methods like
`buildDeleteCondition`, `GetDeleteData`, and improved error logging for
invalid field types.
[[1]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L287-R234)
[[2]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L313-R395)
**Cache Key Generation Enhancements**
- Added dedicated helper functions for generating cache keys for table
names, table fields, select queries, and soft time field/type lookups,
improving cache consistency and code readability.
[[1]](diffhunk://#diff-d57d57e6f9b342ba6fa30c4bb413e2f4f3514a8cd5ad36949eef126e5f8b7ac9R969)
[[2]](diffhunk://#diff-d57d57e6f9b342ba6fa30c4bb413e2f4f3514a8cd5ad36949eef126e5f8b7ac9R980)
[[3]](diffhunk://#diff-d57d57e6f9b342ba6fa30c4bb413e2f4f3514a8cd5ad36949eef126e5f8b7ac9R993-R1002)
[[4]](diffhunk://#diff-b1bbe5e3995261813e4e0ac6ffee8a37c236eaa2759f2bd82e211711695a70bcL790-R790)
**General Code Cleanup**
- Removed redundant code, clarified comments, and improved naming
throughout the affected files, making the code easier to follow and
maintain.
[[1]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L105-R180)
[[2]](diffhunk://#diff-6c1d606032d981a7b8aecd3a7167823f76b69407a29eb9a244175a82f59965d8L255-R206)
Let me know if you'd like a walkthrough of any specific part of the
refactored soft time logic!
When using JOIN queries in MySQL with the `Group()` method, column names
in GROUP BY clauses become ambiguous if multiple tables contain columns
with the same name (commonly `id`). This results in MySQL errors like
"Column 'id' in group statement is ambiguous".
**Example Issue:**
```go
model := t.Ctx(ctx).Fields("t_inf_job.*, t_inf_job_attr.*").
LeftJoin("t_inf_job_attr", "t_inf_job.id = t_inf_job_attr.job_id").
Where(t.Columns().Deleted, 0)
// This would fail with "Column 'id' in group statement is ambiguous"
err = model.Group(t.Columns().Id).Scan(&jobs)
```
### **Key Changes**
1. **Modified function signature**: `Group(groupBy ...string)` →
`Group(groupBy ...any)` to support Raw SQL expressions
2. **Auto-prefixing logic**: When JOINs are detected (by checking for "
JOIN " in the tables string), unqualified column names are automatically
prefixed with the primary table name
3. **Preserved existing behavior**: Already qualified columns
(containing ".") and Raw expressions are handled as before
4. **Added comprehensive test**: `Test_Model_Group_WithJoin` verifies
the fix works correctly with JOIN queries
---------
Co-authored-by: hailaz <739476267@qq.com>
Code example：
``` go
var res *BasicInfo
err := g.DB().Model("basic_info").
FieldsPrefix("basic_info", basicInfoColumns).
Where("id", 35813305356386305).Scan(&res)
if err != nil {
panic(err)
}
g.Dump(res)
```
SQL generated before modification ：
``` sql
SELECT basic_info.id,basic_info.full_name,basic_info.contact FROM `basic_info` WHERE (`id`=35813305356386305) AND `delete_time` IS NULL LIMIT 1
```
SQL generated after modification：
``` sql
SELECT `basic_info`.`id`,`basic_info`.`full_name`,`basic_info`.`contact` FROM `basic_info` WHERE (`id`=35813305356386305) AND `delete_time` IS NULL LIMIT 1
```
---------
Co-authored-by: hailaz <739476267@qq.com>
Change the return type of the GetArray method to Array for consistency
in data structure. Introduce a new Bools method to convert Vars to a
slice of bools, along with corresponding test cases to validate the
functionality.
```go
// 改进前
res, _ := db.Model(table).Fields("id").Array()
ids := make([]int64, 0, len(res))
for _, v := range res {
ids = append(ids, v.Int64())
}
g.Dump(ids)
// 改进后
res, _ := db.Model(table).Fields("id").Array()
ids := res.Int64s()
g.Dump(ids)
```
This pull request standardizes the use of the Go 1.18+ `any` type alias
instead of `interface{}` throughout the codebase. The change improves
code readability and aligns with modern Go best practices. The update
touches many files, including core data structures, code generation
templates, logging utilities, and test data, ensuring consistency across
all usages.
**Type alias migration to `any`:**
* Replaced all instances of `interface{}` with `any` in core data
structures such as `garray` and in generated model structs (e.g.,
`TableUser`, `User1`, `User2`) to modernize type usage.
[[1]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L31-R31)
[[2]](diffhunk://#diff-6c19859cb32c7516ea95ddc8f8235460818eb2f24d2204308e0d9e1b19e7d90fL15-R19)
[[3]](diffhunk://#diff-a15ba2f5e830b4833c47b902515a4f9e5a4f83a3707698f3229b307ec3776b41L15-R18)
[[4]](diffhunk://#diff-52e0837e84d49221d1b810d88fdf78221f36cffcd664fb42f8aba49a79b974dcL15-R19)
[[5]](diffhunk://#diff-11c3457d1a23a4ca6ecd00d6b856289774936b6a708384cf03aff164044e7546L15-R19)
[[6]](diffhunk://#diff-2cff9cf8e6a0cc34087326d8c8149c3bbaf74c76fdbdf5a73daed13cc04249e1L15-R19)
* Updated function signatures, method parameters, and return types from
`interface{}` to `any` in various parts of the codebase, including code
generation, service logic, and logging utilities (e.g., `mlog`).
[[1]](diffhunk://#diff-175edfeea54490b8fe4e18ffcbea5835efaf8f0b8acf623359073987cae7eb76L48-R55)
[[2]](diffhunk://#diff-2b1953fb78cf3593d8c2c7d911e95b65fd0b847c30ed0b4d167d16fe6d781235L54-R74)
[[3]](diffhunk://#diff-e001b7a4b63603b9b14f00de78a4d570bb76c5f57d856a24643f071032e12356L66-R73)
[[4]](diffhunk://#diff-5582954e8a9983988dc8854ad82067fb2ac6269b988e07357ad8db1dfec5f1a0L39-R41)
[[5]](diffhunk://#diff-c5d51d56f487779a2b6207c7ad26c7a20bbadcc846ce094fe60ab4cabff58c51L107-R107)
[[6]](diffhunk://#diff-f96e6a9fdb416eb1804ceaba1fe0ac637bff22c43837f8bb849c2366ce72d4a1L116-R121)
[[7]](diffhunk://#diff-f94c83a1b08ae060d9346f4a6031fc4a7b9a0b894e02d9afaa09018b6598eac0L112-R112)
[[8]](diffhunk://#diff-748b11dbe8828dd4c040ec23cae0b8fe57ecf0a2d1b7694ea39102294e633c64L36-R36)
[[9]](diffhunk://#diff-748b11dbe8828dd4c040ec23cae0b8fe57ecf0a2d1b7694ea39102294e633c64L74-R74)
[[10]](diffhunk://#diff-748b11dbe8828dd4c040ec23cae0b8fe57ecf0a2d1b7694ea39102294e633c64L96-R96)
**Generated code and templates:**
* Adjusted generated files and code generation templates to output `any`
instead of `interface{}` for relevant struct fields and function
signatures, ensuring that new code generation aligns with the updated
convention.
[[1]](diffhunk://#diff-6c19859cb32c7516ea95ddc8f8235460818eb2f24d2204308e0d9e1b19e7d90fL15-R19)
[[2]](diffhunk://#diff-a15ba2f5e830b4833c47b902515a4f9e5a4f83a3707698f3229b307ec3776b41L15-R18)
[[3]](diffhunk://#diff-52e0837e84d49221d1b810d88fdf78221f36cffcd664fb42f8aba49a79b974dcL15-R19)
[[4]](diffhunk://#diff-11c3457d1a23a4ca6ecd00d6b856289774936b6a708384cf03aff164044e7546L15-R19)
[[5]](diffhunk://#diff-2cff9cf8e6a0cc34087326d8c8149c3bbaf74c76fdbdf5a73daed13cc04249e1L15-R19)
[[6]](diffhunk://#diff-175edfeea54490b8fe4e18ffcbea5835efaf8f0b8acf623359073987cae7eb76L48-R55)
[[7]](diffhunk://#diff-e001b7a4b63603b9b14f00de78a4d570bb76c5f57d856a24643f071032e12356L66-R73)
[[8]](diffhunk://#diff-5582954e8a9983988dc8854ad82067fb2ac6269b988e07357ad8db1dfec5f1a0L39-R41)
**Container and utility updates:**
* Refactored the `garray` container implementation and related
constructors/methods to use `[]any` instead of `[]interface{}`, along
with corresponding function signatures.
[[1]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L31-R31)
[[2]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L52-R52)
[[3]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L62-R62)
[[4]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L73-R86)
[[5]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L96-R97)
[[6]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L107-R114)
[[7]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L124-R124)
[[8]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L135-R143)
[[9]](diffhunk://#diff-3a1259e160a4dfa5fe49dfe739fbdb986c0d0a2220a709882ea48d3ae1b8f911L167-R167)
These changes collectively modernize the codebase and prepare it for
future Go developments by using the idiomatic `any` type.
* add Tag* functions to retreive most commonly used tag value from struct field for package gstructs; use description tag as default value if brief is empty for gcmd.Argument
* fix cache issue in Count/Value functions for gdb.Model
* add more ut case for package gdb
* version updates
