Speed up startup

providers/agents.go

@@ -11,11 +11,11 @@ type AgentProvider struct {
 
 type Agent struct {
 	XMLName  xml.Name `xml:"akteur"`
-	ID       string   `xml:"id,attr"`
 	Names    []string `xml:"name"`
 	SortName string   `xml:"sortiername"`
 	Life     string   `xml:"lebensdaten"`
 	GND      string   `xml:"gnd"`
+	Identifier
 	AnnotationNote
 }
 

providers/categories.go

@@ -15,9 +15,10 @@ type Categories struct {
 }
 
 type Category struct {
-	ID       string   `xml:"id,attr"`
+	XMLName  xml.Name `xml:"kategorie"`
 	Names    []string `xml:"name"`
 	SortName string   `xml:"sortiername"`
+	Identifier
 	AnnotationNote
 }
 

providers/git.go

@@ -13,9 +13,12 @@ import (
 
 var InvalidBranchError = errors.New("The currently checked out branch does not match the requested branch. Please checkout the correct branch first.")
 var InvalidStateError = errors.New("The GitProvider is not in a valid state. Fix the issues or continue without Git data.")
+var NoURLProvidedError = errors.New("No URL provided for GitProvider.")
+var NoPathProvidedError = errors.New("No path or branch provided for GitProvider.")
 
-// NOTE: GitProvider does not open the files, it can only
-// - clone the repo, given an URL
+// NOTE: GitProvider does not open XML files, it can only
+// - read in information from the repo, given a path
+// - clone the repo, given an URL & a path
 // - pull the repo, given a path
 // In case of success in either case it updates the commit hash and date and closes the repo again.
 // The Files are opened  and serialized by the FSProvider, which operates on the same file path.
@@ -29,33 +32,67 @@ type GitProvider struct {
 	Date   time.Time
 }
 
-func NewGitProvider(url string, path string, branch string) *GitProvider {
-	if branch == "" || url == "" || path == "" {
-		return nil
+func NewGitProvider(url string, path string, branch string) (*GitProvider, error) {
+	// TODO: check if directory is empty
+	// TODO: force clone
+	if _, err := os.Stat(path); err == nil {
+		return GitProviderFromPath(path, branch)
 	}
-	return &GitProvider{URL: url, Path: path, Branch: branch}
+
+	return GitProviderFromURL(url, path, branch)
 }
 
-func (g *GitProvider) Pull() error {
+func GitProviderFromPath(path string, branch string) (*GitProvider, error) {
+	if branch == "" || path == "" {
+		return nil, NoPathProvidedError
+	}
+
+	gp := GitProvider{Path: path, Branch: branch}
+	if err := gp.Read(); err != nil {
+		return nil, err
+	}
+
+	return &gp, nil
+}
+
+func GitProviderFromURL(url string, path string, branch string) (*GitProvider, error) {
+	if url == "" {
+		return nil, NoURLProvidedError
+	}
+
+	if branch == "" || path == "" {
+		return nil, NoPathProvidedError
+	}
+
+	gp := GitProvider{URL: url, Path: path, Branch: branch}
+	if err := gp.Clone(); err != nil {
+		return nil, err
+	}
+
+	return &gp, nil
+}
+
+// Returs true if the repo was updated remotely, false otherwise
+func (g *GitProvider) Pull() (error, bool) {
 	g.mu.Lock()
 	defer g.mu.Unlock()
 
 	branch := plumbing.NewBranchReferenceName(g.Branch)
 	repo, err := git.PlainOpen(g.Path)
 	if err != nil {
-		return err
+		return err, false
 	}
 
 	wt, err := repo.Worktree()
 	if err != nil {
-		return err
+		return err, false
 	}
 
 	if err := wt.Checkout(&git.CheckoutOptions{
 		Branch: branch,
 		Force:  true,
 	}); err != nil {
-		return err
+		return err, false
 	}
 
 	if err := wt.Pull(&git.PullOptions{
@@ -63,14 +100,20 @@ func (g *GitProvider) Pull() error {
 		ReferenceName: branch,
 		Progress:      os.Stdout,
 	}); err != nil && err != git.NoErrAlreadyUpToDate {
-		return err
+		return err, false
+	} else if err == git.NoErrAlreadyUpToDate {
+		return nil, false
 	}
 	defer wt.Clean(&git.CleanOptions{Dir: true})
 
-	return g.setValues(repo)
+	return g.setValues(repo), true
 }
 
 func (g *GitProvider) Clone() error {
+	if g.URL == "" {
+		return NoURLProvidedError
+	}
+
 	g.mu.Lock()
 	defer g.mu.Unlock()
 
@@ -106,7 +149,7 @@ func (g *GitProvider) Clone() error {
 
 // Implement String Interface
 func (g *GitProvider) String() string {
-	return fmt.Sprintf("GitProvider{URL: %s, Path: %s, Branch: %s, Commit: %s, Date: %s}", g.URL, g.Path, g.Branch, g.Commit, g.Date)
+	return fmt.Sprintf("GitProvider\nURL: %s\nPath: %s\nBranch: %s\nCommit: %s\nDate: %s\n", g.URL, g.Path, g.Branch, g.Commit, g.Date)
 }
 
 func (g *GitProvider) setValues(repo *git.Repository) error {
@@ -124,10 +167,9 @@ func (g *GitProvider) setValues(repo *git.Repository) error {
 	g.Commit = commit.Hash.String()
 	g.Date = commit.Author.When
 
-	return err
+	return nil
 }
 
-// WARNING: this expects the repo to be in a certain state and is intended to be used in tests.
 func (g *GitProvider) Read() error {
 	g.mu.Lock()
 	defer g.mu.Unlock()
@@ -137,6 +179,47 @@ func (g *GitProvider) Read() error {
 		return err
 	}
 
+	if err := g.ValidateBranch(repo); err != nil {
+		branch := plumbing.NewBranchReferenceName(g.Branch)
+		wt, err := repo.Worktree()
+		if err != nil {
+			return err
+		}
+		defer wt.Clean(&git.CleanOptions{Dir: true})
+
+		if err := wt.Checkout(&git.CheckoutOptions{
+			Branch: branch,
+			Force:  true,
+		}); err != nil {
+			return err
+		}
+
+		if err := g.ValidateBranch(repo); err != nil {
+			return err
+		}
+	}
+
+	return g.setValues(repo)
+}
+
+func (g *GitProvider) Validate() error {
+	repo, err := git.PlainOpen(g.Path)
+	if err != nil {
+		return err
+	}
+
+	if err := g.ValidateBranch(repo); err != nil {
+		return err
+	}
+
+	if err := g.ValidateCommit(); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func (g *GitProvider) ValidateBranch(repo *git.Repository) error {
 	head, err := repo.Head()
 	if err != nil {
 		return err
@@ -147,13 +230,10 @@ func (g *GitProvider) Read() error {
 		return InvalidBranchError
 	}
 
-	return g.setValues(repo)
+	return nil
 }
 
-func (g *GitProvider) Validate() error {
-	g.mu.Lock()
-	defer g.mu.Unlock()
-
+func (g *GitProvider) ValidateCommit() error {
 	if g.Commit == "" || g.Date.IsZero() {
 		return InvalidStateError
 	}

providers/issues.go

@@ -15,26 +15,30 @@ type Issues struct {
 }
 
 type Issue struct {
-	ID          string       `xml:"id,attr"`
+	XMLName     xml.Name     `xml:"stueck"`
 	Number      IssueNumber  `xml:"nummer"`
 	Datum       KGPZDate     `xml:"datum"`
 	Von         string       `xml:"von"`
 	Bis         string       `xml:"bis"`
 	Additionals []Additional `xml:"beilage"`
+	Identifier
+	AnnotationNote
 }
 
 type IssueNumber struct {
-	XMLName   xml.Name `xml:"nummer"`
-	Value     string   `xml:",chardata"`
-	Corrected string   `xml:"korrigiert,attr"`
+	XMLName xml.Name `xml:"nummer"`
+	Value
+	Corrected string `xml:"korrigiert,attr"`
 }
 
 type KGPZDate struct {
-	When      string `xml:"when,attr"`
-	NotBefore string `xml:"notBefore,attr"`
-	NotAfter  string `xml:"notAfter,attr"`
-	From      string `xml:"from,attr"`
-	To        string `xml:"to,attr"`
+	XMLName   xml.Name `xml:"datum"`
+	When      string   `xml:"when,attr"`
+	NotBefore string   `xml:"notBefore,attr"`
+	NotAfter  string   `xml:"notAfter,attr"`
+	From      string   `xml:"from,attr"`
+	To        string   `xml:"to,attr"`
+	Value
 }
 
 type Additional struct {

providers/pieces.go

@@ -15,7 +15,7 @@ type Pieces struct {
 }
 
 type Piece struct {
-	ID            string          `xml:"id,attr"`
+	XMLName       xml.Name        `xml:"beitrag"`
 	IssueRefs     []IssueRef      `xml:"stueck"`
 	PlaceRefs     []PlaceRef      `xml:"ort"`
 	CategoryRefs  []CategoryRef   `xml:"kategorie"`
@@ -25,50 +25,10 @@ type Piece struct {
 	AdditionalRef []AdditionalRef `xml:"beilage"`
 	Incipit       []string        `xml:"incipit"`
 	Title         []string        `xml:"titel"`
+	Identifier
 	AnnotationNote
 }
 
-type AdditionalRef struct {
-	Datum string `xml:"datum,attr"`
-	Nr    string `xml:"nr,attr"`
-	Von   string `xml:"von,attr"`
-	Bis   string `xml:"bis,attr"`
-	Value string `xml:",chardata"`
-}
-
-type IssueRef struct {
-	Datum    string `xml:"datum,attr"`
-	Nr       string `xml:"nr,attr"`
-	Von      string `xml:"von,attr"`
-	Bis      string `xml:"bis,attr"`
-	Value    string `xml:",chardata"`
-	Category string `xml:"kat,attr"`
-}
-
-type PlaceRef struct {
-	Ref      string `xml:"ref,attr"`
-	Value    string `xml:",chardata"`
-	Category string `xml:"kat,attr"`
-}
-
-type CategoryRef struct {
-	Ref   string `xml:"ref,attr"`
-	Value string `xml:",chardata"`
-}
-
-type WorkRef struct {
-	Ref      string `xml:"ref,attr"`
-	Value    string `xml:",chardata"`
-	Category string `xml:"kat,attr"`
-	Page     string `xml:"s,attr"`
-}
-
-type PieceRef struct {
-	Ref      string `xml:"ref,attr"`
-	Category string `xml:"kat,attr"`
-	Value    string `xml:",chardata"`
-}
-
 func (p Pieces) Append(data Pieces) Pieces {
 	p.Piece = append(p.Piece, data.Piece...)
 	return p

providers/places.go

@@ -15,10 +15,11 @@ type Places struct {
 }
 
 type Place struct {
-	ID       string   `xml:"id,attr"`
+	XMLName  xml.Name `xml:"ort"`
 	Names    []string `xml:"name"`
 	SortName string   `xml:"sortiername"`
 	Geo      string   `xml:"geonames"`
+	Identifier
 	AnnotationNote
 }
 

providers/works.go

@@ -15,24 +15,14 @@ type Works struct {
 }
 
 type Work struct {
-	ID       string     `xml:"id,attr"`
+	XMLName  xml.Name   `xml:"werk"`
 	URLs     []URL      `xml:"url"`
 	Citation []string   `xml:"zitation"`
 	Akteur   []AgentRef `xml:"akteur"`
+	Identifier
 	AnnotationNote
 }
 
-type AgentRef struct {
-	Ref      string `xml:"ref,attr"`
-	Category string `xml:"Kat,attr"`
-	Value    string `xml:",chardata"`
-}
-
-type URL struct {
-	Address string `xml:"address,attr"`
-	Value   string `xml:",chardata"`
-}
-
 func (w Works) Append(data Works) Works {
 	w.Work = append(w.Work, data.Work...)
 	return w

providers/xmlcommon.go

@@ -1,48 +1,55 @@
 package providers
 
-import (
-	"encoding/xml"
-	"fmt"
-	"io"
-	"os"
-	"sync"
-)
+import "encoding/xml"
 
-type KGPZXML[T any] interface {
-	Append(data T) T
-	fmt.Stringer
+type AgentRef struct {
+	XMLName xml.Name `xml:"akteur"`
+	Reference
 }
 
-type XMLProvider[T KGPZXML[T]] struct {
-	mu    sync.Mutex
-	paths []string
-	Items T
+type URL struct {
+	XMLName xml.Name `xml:"url"`
+	Address string   `xml:"address,attr"`
+	Value
 }
 
-func (p *XMLProvider[T]) Load() error {
-	var wg sync.WaitGroup
-	for _, path := range p.paths {
-		wg.Add(1)
-		go func(path string) {
-			defer wg.Done()
-			var data T
-			if err := UnmarshalFile(path, &data); err != nil {
-				fmt.Println(err)
-				return
-			}
-			p.mu.Lock()
-			p.Items = p.Items.Append(data)
-			p.mu.Unlock()
-		}(path)
-	}
-	wg.Wait()
-	return nil
+type AdditionalRef struct {
+	XMLName xml.Name `xml:"beilage"`
+	Reference
+	Datum string `xml:"datum,attr"`
+	Nr    string `xml:"nr,attr"`
+	Von   string `xml:"von,attr"`
+	Bis   string `xml:"bis,attr"`
 }
 
-func (a *XMLProvider[T]) String() string {
-	a.mu.Lock()
-	defer a.mu.Unlock()
-	return fmt.Sprintf("Items: %s", a.Items)
+type IssueRef struct {
+	XMLName xml.Name `xml:"stueck"`
+	Reference
+	Datum string `xml:"datum,attr"`
+	Nr    string `xml:"nr,attr"`
+	Von   string `xml:"von,attr"`
+	Bis   string `xml:"bis,attr"`
+}
+
+type PlaceRef struct {
+	XMLName xml.Name `xml:"ort"`
+	Reference
+}
+
+type CategoryRef struct {
+	XMLName xml.Name `xml:"kategorie"`
+	Reference
+}
+
+type WorkRef struct {
+	XMLName xml.Name `xml:"werk"`
+	Reference
+	Page string `xml:"s,attr"`
+}
+
+type PieceRef struct {
+	XMLName xml.Name `xml:"beitrag"`
+	Reference
 }
 
 type AnnotationNote struct {
@@ -50,16 +57,17 @@ type AnnotationNote struct {
 	Notes       []string `xml:"vermerk"`
 }
 
-func UnmarshalFile[T any](filename string, data *T) error {
-	xmlFile, err := os.Open(filename)
-	if err != nil {
-		fmt.Println(err)
-		return err
-	}
-	fmt.Println("Successfully opened " + filename)
-	defer xmlFile.Close()
-	byteValue, _ := io.ReadAll(xmlFile)
-	xml.Unmarshal(byteValue, data)
-
-	return nil
+type Identifier struct {
+	ID string `xml:"id,attr"`
+}
+
+type Reference struct {
+	Ref      string `xml:"ref,attr"`
+	Category string `xml:"kat,attr"`
+	Unsicher bool   `xml:"unsicher,attr"`
+	Value
+}
+
+type Value struct {
+	Value string `xml:",chardata"`
 }

providers/xmlprovider.go

@@ -0,0 +1,60 @@
+package providers
+
+import (
+	"encoding/xml"
+	"fmt"
+	"io"
+	"os"
+	"sync"
+)
+
+type KGPZXML[T any] interface {
+	Append(data T) T
+	fmt.Stringer
+}
+
+type XMLProvider[T KGPZXML[T]] struct {
+	mu    sync.Mutex
+	paths []string
+	Items T
+}
+
+func (p *XMLProvider[T]) Load() error {
+	var wg sync.WaitGroup
+	for _, path := range p.paths {
+		wg.Add(1)
+		go func(path string) {
+			defer wg.Done()
+			var data T
+			if err := UnmarshalFile(path, &data); err != nil {
+				fmt.Println(err)
+				return
+			}
+			p.mu.Lock()
+			p.Items = p.Items.Append(data)
+			p.mu.Unlock()
+		}(path)
+	}
+	wg.Wait()
+	return nil
+}
+
+func (a *XMLProvider[T]) String() string {
+	a.mu.Lock()
+	defer a.mu.Unlock()
+	return fmt.Sprintf("Items: %s", a.Items)
+}
+
+func UnmarshalFile[T any](filename string, data *T) error {
+	xmlFile, err := os.Open(filename)
+	if err != nil {
+		fmt.Println(err)
+		return err
+	}
+	fmt.Println("Successfully opened " + filename)
+	defer xmlFile.Close()
+	byteValue, _ := io.ReadAll(xmlFile)
+	xml.Unmarshal(byteValue, data)
+
+	return nil
+}